FEATURES
The Clavien-Dindo Classification of Surgical Complications Five-Year Experience Pierre A. Clavien, MD, PhD,* Jeffrey Barkun, MD,† Michelle L. de Oliveira, MD, PhD,* Jean Nicolas Vauthey, MD,‡ Daniel Dindo, MD,* Richard D. Schulick, MD,§ Eduardo de Santiban˜es, MD, PhD,¶ Juan Pekolj, MD, PhD,¶ Ksenija Slankamenac, MD,* Claudio Bassi, MD,储 Rolf Graf, PhD,* Rene´ Vonlanthen, MD,* Robert Padbury, MD, PhD,** John L. Cameron, MD,§ and Masatoshi Makuuchi, MD, PhD††
Background and Aims: The lack of consensus on how to define and grade adverse postoperative events has greatly hampered the evaluation of surgical procedures. A new classification of complications, initiated in 1992, was updated 5 years ago. It is based on the type of therapy needed to correct the complication. The principle of the classification was to be simple, reproducible, flexible, and applicable irrespective of the cultural background. The aim of the current study was to critically evaluate this classification from the perspective of its use in the literature, by assessing interobserver variability in grading complex complication scenarios and to correlate the classification grades with patients’, nurses’, and doctors’ perception. Material and Methods: Reports from the literature using the classification system were systematically analyzed. Next, 11 scenarios illustrating difficult cases were prepared to develop a consensus on how to rank the various complications. Third, 7 centers from different continents, having routinely used the classification, independently assessed the 11 scenarios. An agreement analysis was performed to test the accuracy and reliability of the classification. Finally, the perception of the severity was tested in patients, nurses, and physicians by presenting 30 scenarios, each illustrating a specific grade of complication. Results: We noted a dramatic increase in the use of the classification in many fields of surgery. About half of the studies used the contracted form, whereas the rest used the full range of grading. Two-thirds of the publications avoided subjective terms such as minor or major complications. The study of 11 difficult cases among various centers revealed a high degree of agreement in identifying and ranking complications (89% agreement), and enabled a better definition of unclear situations. Each grade of complications significantly correlated with the perception by patients, nurses, and physicians (P ⬍ 0.05, Kruskal-Wallis test). Conclusions: This 5-year evaluation provides strong evidence that the classification is valid and applicable worldwide in many fields of surgery. No modification in the general principle of classification is warranted in view of the use in ongoing publications and trials. Subjective, inaccurate, or confus-
From the *Department of Surgery and Swiss HPB Center, University Hospital of Zurich, Switzerland; †Department of Surgery, McGill University, Montreal, Canada; ‡Department of Surgery, MD Anderson Cancer Center, Houston, TX; §Department of Surgery, Johns Hopkins Medical Institutions, Baltimore, MD; ¶Department of Surgery, Hospital Italiano, Buenos Aires, Argentina; 㛳Department of Surgery, Borgo Roma University Hospital, Verona, Italy, **Department of Surgery and Specialty Services, Flinders Medical Centre, Adelaide, Australia; and ††Department of Surgery, Red Cross Hospital, Tokyo, Japan. Reprints: Pierre A. Clavien, MD, PhD, Department of Surgery, University Hospital of Zurich, Ramistrasse 100, 8091 Zurich, Switzerland. E-mail:
[email protected]. Supplemental digital content is available for this article. Direct URL citations appear in the printed text, and links to the digital files are available in the HTML text of this article on the journal’s Web site (www.annalsofsurgery.com). Copyright © 2009 by Lippincott Williams & Wilkins ISSN: 0003-4932/09/25002-0187 DOI: 10.1097/SLA.0b013e3181b13ca2
Annals of Surgery • Volume 250, Number 2, August 2009
ing terms such as “minor or major” should be removed from the surgical literature. (Ann Surg 2009;250: 187–196)
T
he absence of a definition and a widely accepted ranking system to classify surgical complications has hampered proper interpretation of surgical outcome data for a long time.1 Terms, such as minor, moderate, major, or severe complications, have been inconsistently used among authors, centers, and over time periods.2 A number of attempts have been made in the 1990s to classify surgical complications,2– 6 but none of them have gained widespread acceptance. In 1992, a novel approach was presented to rank complications by severity based on the therapy used to treat the complications, and differentiated 3 types of negative outcome after surgery, (a) complication, (b) failure to cure, and (c) sequela.2 Although this system was used by few investigators, we revisited this grading system in 2004, after its routine use for more than 12 years. We developed a new 5-scale classification system with the aim of presenting an objective, simple, reliable, and reproducible way of reporting negative events after surgery7 (Appendix A). This classification was further validated before publication through a large cohort of patients, who underwent a variety of surgical procedures. The new proposal was also tested for its simplicity and “interobserver” variation in 10 centers around the world. Similar to the initial classification,2 this new system7 was based on the type of therapy required to treat the complication. The rationale to preserve this approach was to eliminate subjective interpretation of serious adverse events and any tendency to down-grade complications, because it is based on data that are usually well documented and easily verified. To further avoid subjectivity and imprecision in complication reporting, we purposely avoided qualitative terms such as “minor” or “major” to grade the complications. Compared with the 1992 system,2 we also eliminated hospital stay as a criterion and increased the weight of life-threatening complications involving organ failure. We gave more emphasis to the patient perspective by introducing the notion of disability indicating the need for further follow-up, which could be added to each type of complication. Finally, the classification offered the possibility to combine grades of complications to simplify its use depending on the focus and the patient cohort that is analyzed. This classification has been used in many centers as a tool for quality assessment in audits and every day practice, and it is increasingly used in the surgical literature.8 –21 It has also been endorsed by societies and study investigators; for example, by the Transplantation Society to record living-related liver transplantation in the United States.22 We are also aware of many randomized trials, which have included this system in their end points. Every complication has been recorded in the prospective quality database at the University Hospital of Zurich since the beginning of its application, www.annalsofsurgery.com | 187
Annals of Surgery • Volume 250, Number 2, August 2009
Clavien et al
more than 5 years ago. Each complication is mentioned in the weekly morbidity and mortality (M&M) conference, where selected cases are also discussed in more detail. From this conference each case was recorded prospectively. Some cases triggered controversial discussions regarding the grading of severity. The wide implementation in clinical practice and studies of such a system requires time. Therefore, changes, unless clearly justified, are unwarranted, since they may only lead to confusion and inconsistencies in reporting outcome. We initiated a 5-year critical evaluation of this system to identify whether it had reached its goal, and whether or not substantial changes were needed. To achieve this goal, we performed an analysis from 4 points of view. First, we searched for all articles, in which this system was cited and to identify how the complication system was used. Second, the University of Zurich authors reviewed each case, which had previously generated controversies at their weekly M&M conferences. Third, a total of 7 centers covering 5 continents, which had used the system routinely for more than 3 years, were contacted to interpret these controversial cases. Finally, we tested how the severity of each grade is perceived among 3 groups of stake holders (physicians, nurses, and patients) through case presentations.
METHODS First Part: Analysis of the Literature We used the Institute of Scientific Information Web of Science (Available at: http://apps.isiknowledge.com), an online database for citations administrated by the Institute of Scientific Information. In this database, we searched for published articles that have referenced our initial publication of 2004. All articles were considered, including original articles, editorials, and reviews. Each original article was reviewed to assess how the classification was applied in terms of contraction of grades (grade III ⫽ IIIa ⫹ IIIb and grade IV ⫽ IVa ⫹ IVb) and any modifications that have been undertaken.
each grade of complication, except death (grade V). The descriptions of the cases were adjusted for the audience, respectively for physicians, nurses, and patients (case description available on demand). These 3 populations (n ⫽ 50 in each category) were asked to grade each scenario on a numerical visual analogue scale from 0 (not severe) to 100 (extremely severe). None of the participants was aware of the respective grade allocated to each scenario, and were not informed about the concept of the grading system. We calculated the perception of severity of surgical complications between 3 categories with the nonparametric test for multivariate analysis of variance (Kruskal-Wallis test) with Bonferroni correction. The level of statistical significance was P ⬍ 0.05.
RESULTS First Part: Analysis of the Literature A: Use of the Complication Classification in the Literature The first part of our evaluation focused on the use of the classification in the literature. From the publication of the original article in 2004 until March 2009, we identified 214 articles, which cited or used the complication system. The complete list of references is available on the website (Available at: www.surgicalcomplication.info). Our analysis focused on how frequently the classification was used (Fig. 1), in which field of medicine, and for what type (methodology) of study. As one key feature of the new system was the option to regroup (“contract”) subtypes of complications, we also looked at whether the “standard” or the “contracted” forms of the complication system were applied. Finally, as the complication
Second Part: Prospective Evaluation of “Difficult” Cases During the past 5 years, authors form the University of Zurich have collected a number of clinical cases from their weekly M&M conferences, which have provoked a discussion on the uncertainties around the application of the grading system. To illustrate those cases, a number of scenarios have been prepared, and will be presented with the consensus decision on the interpretation of the respective negative events.
Third Part: Assessment of the Complication Grading in Seven Centers From Each Continent The wide implementation of a reporting system mostly depends on its acceptance and use in other centers. To look at this aspect, 7 centers from around the world, where the classification has been routinely applied for more than 3 years, received 11 difficult cases for interpretation. They were also asked to submit cases presenting possible controversial interpretation of a negative event. In addition, we calculated the concordance among centers. To support the understanding of our complication grading we implemented a webpage, which defined complication and provide forms to record the grades (Available at: www.surgicalcomplication.info).
Fourth Part: Correlation Between Severity of Each Respective Grade of Postoperative Complications and Perception Among Patients, Nurses, and Physicians To further elucidate the objectivity and utility of this scoring system we prepared 30 cases, each illustrating a frequent, relevant, and specific postoperative complication. The 30 scenarios described 188 | www.annalsofsurgery.com
FIGURE 1. Citations of the classification since its introduction in the literature in 2004 (f) (A). The classification experienced a steady in use, while the inaugural system proposed in 1992 (䡺) (B), which constituted the basis for the new scale, failed to gain wide acceptance. *Data for 2009 include only the months from January to March. © 2009 Lippincott Williams & Wilkins
Annals of Surgery • Volume 250, Number 2, August 2009
system was originally presented omitting a specific name (classification of surgical complications, a new proposal…),7 we were interested in the terminology used in the literature with the aim to suggest a standard name to, thereby, enable consistent reporting in the literature. We documented a steady yearly increase in the citation of the classification since its introduction in 20047 (Fig. 1A), which is in contrast to the citation of the inaugural classification system available since 19922 (Fig. 1B). Of the 214 citations of the classification system until March 2009, 161 (75%) were from original articles, in which the classification was applied to record and grade complications. All articles originate from surgical specialties, including 31 (19%) articles from the transplantation field, 49 (31%) from the hepato-pancreatico-biliary field, 26 (16%) from general surgery, 23 (14%) from colorectal surgery, 23 (14%) from urology, and finally 9 (6%) from gynecology. Those original articles included 11 randomized trials (7%), 7 prospective (4%), 6 (3%) matched-control, and 137 (85%) retrospective studies. Of note, 22 of the 30 nonoriginal articles, mostly reviews focusing on surgical complications23–26 recommended the routine use of the classification of complications to report outcome. Contracted and Complete Forms Versus Modification. Interestingly, of 161 original reports, 79 (49%) used the contracted form as suggested in the original article7 (adding together the scores “a” and “b” for grades III and IV each).14,17,27– 40 There was no correlation between the use of the contracted form and the number of patients included in the study. For example, 45 publications (57%) presenting the contracted form included more than 100 patients, whereas 34 (43%) publications had less than 100 patients. The remaining 82 original studies (51%) have recorded the full range of grades and subgrades,21,41– 49 although 30% (n ⫽ 25) reported only selective data in the results section.28 –31 A number of changes were also suggested; for example, one group excluded the need for blood transfusion in the postoperative period as a complication,10,11 while another group omitted recording grade I complications.32 Minor, Moderate, and Major Complications. Most of the original reports (n ⫽ 112, 63%) did not use terms such as “minor or major complications” in any parts of the article, which is consistent with the recommendations made in the original report.7 Of the remaining studies (n ⫽ 45, 37%) the term minor was mostly used to refer to grades I and II complications, whereas major related to grades III to V complications.15,33,34 The MD Anderson group applied the term minor for grades I and II complications, major for grades III and IV, and, separately, reported death (grade V complication).35,36 Seven percent of the authors (n ⫽ 4) made different choices; for example, considering minor complications as grades I to IIIa and major as grades IIIb to V,50 –52 or the term minor for grades I and II, moderate for grade IIIa, and major for grades IIIb to V. Other modifications were adopted contributing to further inconsistencies. Nomenclature of the Classification. The denomination of the classification was also inconsistently used among authors. Forty-four percent (n ⫽ 94) of the all articles13,14,16 –19,22–26,53–93 considered this system as a revision of the inaugural article reported in 1992 by Clavien et al,2 and thereby denominated the classification using this name (Clavien classification, system or score, modified, or revised Clavien classification, or Clavien complication scale or scheme). Six percent (n ⫽ 13) labeled it as “Dindo” classification or system,94 –101 and 3% (n ⫽ 6) adopted other terms, such as “updated conventional classification system,” “Zurich classification,” “Toronto modified classification,” and “grade of morbidity.”102–106 In about half of the publications (n ⫽ 82, 47%), only the label “classification of complication” with the proper reference was used.10,11,15,20,28,35,36,107–129 © 2009 Lippincott Williams & Wilkins
Grading of Complications
Second Part: Prospective Evaluation of Difficult Cases The second part of the study focused on reviewing each case presented at the M&M conference at the University of Zurich, which was considered “borderline.” These cases were prospectively recorded over the past 5 years for the purpose of the present analysis. To cover all those cases, 11 scenarios were prepared by the Zurich authors (Appendix B, http://links.lww.com/SLA/A3), and their interpretation is summarized below, and in Table 1.
Scenario 1: Complication After a Surgical Procedure Not Caused by a Surgeon This scenario describes a postoperative complication that is not caused by the surgical team responsible for the patient; this may include complications after an intervention by the radiology or gastroenterology team. Although the “causative” physician is not a surgeon, the complication would not have occurred if the patient had not undergone surgery. Therefore, such postoperative events should be recorded irrespective of the team involved in caring for the patient. The correct labeling of both complications presented in this scenario was biloma (grade IIIa) and pneumothorax (grade IIIa).
Scenario 2: Complication Occurring in a Patient Transferred After a Surgical Procedure Performed in Another Center This patient was referred to a surgical unit from another hospital due to a complicated postoperative course. Typically, such patients are in poor conditions, yielding a high risk of developing additional problems. The surgical unit may be tempted to omit recording such events because the initial surgery including a rough postoperative course, originates from another team. The consensus at the University of Zurich was that all complications must be properly recorded. However, for a quality assessment of the surgical unit, the patients should be identified as transferred from another unit. This team used the addendum “referred patient” in brackets after the complication grade to indicate that the patient was initially not treated in the center reporting the outcome. The grading in this case was grade V (referred patient).
Scenario 3: Patients Developing Complications of Increasing Severity Depending One From Each Other This scenario illustrates cases with the sequential development of a complication gradually becoming more severe. Another example would be a reoperation of a small bowel obstruction, associated with a bronchoaspiration of gastric content and, eventually, death. The question is whether each grade of the same line of complications must be recorded separately, or only the most severe ones. The University of Zurich team records only the most severe grade, when a complication clearly occurs as consequence from a prior, less severe complication. In this case, a grade I (wound infection), a grade II (deep vein thrombosis), and a grade V complication (anastomotic dehiscence leading to death) were recorded because those complications do not depend on each another.
Scenario 4: Complication Still Present at the Time of Discharge or at Follow-Up Visits In this case, the patient remains in a compromised health condition at the time of discharge, as a result of his shaky postoperative course and long hospital stay. The suffix “d” (disabling) was introduced in the classification to catch persistent complications at the time of discharge or at follow-up visits, which often raised the question which of the conditions may qualify for this suffix. A frail www.annalsofsurgery.com | 189
Clavien et al
health status after a long course does not qualify for the label d. Conditions should only be labeled as such, if they describe a specific complication. For example, paresis of a hand after surgery, persisting wound dehiscence etc.
Scenario 5: Intraoperative Death of the Patient This scenario describes an intraoperative death. The classification of 2004 was developed to record postoperative complications, recognizing that intraoperative complications, for example, a bleeding or a tear on the small bowel, is often difficult to define, and, therefore, is unlikely to be reported unless it leads to postoperative negative events. The team in Zurich had, however, considered that death of a patient during surgery must be an exception, and must be caught by the classification. Hence, a grade V complication was recorded in such a scenario.
Scenario 6: Death of a Patient Before Surgical Intervention In this scenario, the patient dies during the intubation attempt, that is, before skin incision. Here, the question arises of whether a complication that occurs before a surgical intervention should be recorded as a surgical complication. In our opinion, each complication that occurs in a surgical patient has to be recorded in an “intention-to-treat” manner. Complication in the preparation for surgery, once the indication is established, should, therefore, be recorded irrespective of whether surgery has been performed or not. This scenario therefore describes a grade V complication.
Scenario 7: Death of a Patient After an Operation Performed in Another Hospital Similarly to case 2, the focus, here, is on a lethal outcome after the transfer of a patient operated elsewhere. In such a situation, the fatal outcome must be recorded in both centers, and the patient should again be properly labeled as “initially operated elsewhere” (grade V, referred patient).
Scenario 8: Patient Undergoing an Explorative “Blank” (Negative Laparotomy) This is a relatively common scenario, in which the suspicion of an abdominal complication justified a laparotomy; but, the laparotomy is unrevealing and the patient subsequently recovers uneventfully. Should such a laparotomy be graded as IIIb complication or not be considered as a complication at all? Although such an intervention was a consequence of the former surgical intervention (aortic aneurysm repair), it was purely diagnostic. As the basis of the complication classification is the therapy required to treat a complication, blank laparotomy should not be considered a complication. This is also important to prevent proper management of the patient, as it might become tempting to avoid an invasive diagnostic procedure to minimize the reported morbidity.
Scenario 9: Patient Developing a Medical Problem Not Obviously Related to Surgery This scenario illustrates negative events occurring in the course of a surgical intervention that are not or only remotely related, to surgery. The correlation between such an event and surgery is, however, sometimes difficult to assess. In this case, the duodenal ulcer might or might not be due to the former surgical intervention. To prevent subjective interpretation, the Zurich team considered that any negative event occurring in a patient during hospital stay or within 30 days after surgery (whichever is longer), should be regarded as a surgical complication and should, therefore, be recorded; the labeling was therefore “grade II.” 190 | www.annalsofsurgery.com
Annals of Surgery • Volume 250, Number 2, August 2009
Scenario 10: Complications After Procedures Performed in Other Departments This case presents a patient that was initially operated in the gynecology service, and then transferred to the surgical ward for a complicated course. An iatrogenic perforation of the colon occurred during the initial operation, which required further management by the surgical team. This is another scenario sharing mechanisms similar to cases 2 and 7. Such complications must be recorded irrespective of the origin or location of the patient; hence, the Zurich team recorded it as a grade IIIb (referred patient) in the displayed scenario.
Scenario 11: Patient With a Medical Problem Not Related to Surgery In this scenario, the patient develops a medical problem unrelated to the previous surgery (incarcerated hernia after thyroid surgery). Despite the doubtful correlation, this scenario was evaluated with the same principle as in case 7 and considered it a grade IIIb complication. The relationship between surgery and an adverse medical event is often speculative and relies on subjective interpretation. Therefore, we reached a consensus to record all the negative events, which have occurred in a patient after surgery, irrespective of whether there was or was not a clear correlation with a former surgical intervention.
Third Part: Assessment of the Complication Grading in Seven Centers From Each Continent Seven centers from around the world, who have routinely used the classification, at least over the past 3 years and have published their results using this system (in at least 1 peer reviewed Journal), were asked to evaluate the 11 scenarios presented earlier. This included Masatoshi Makuuchi, Japan, Eduardo de Santiban˜es and Juan Pekolj, Argentina, Robert Padbury, Australia, Claudio Bassi, Italy, Nicolas Vauthey, United States, Richard Schulick and John Cameron, United States, and Jeffrey Barkun, Canada. The scenarios were mailed to them for their interpretation, without providing any insight about our comments. We also asked for other borderline cases from their institution. An overview of their interpretations is presented in Table 1. The authors from Japan and Australia scored the 11 scenarios similarly to the Zurich team representing an agreement of 100%, while the 5 other centers pointed out some discordance in selective cases, but the rating disclosed 89% agreement. Scenario 3 (patients developing complications of increasing severity depending one from each other) triggered some divergence in interpretation among the centers. Most authors agreed to score only the most severe complication, when clearly depending one from each other. The authors from Japan, Australia, Italy, and Switzerland were in agreement, whereas the authors from Argentina and Baltimore, recorded all events separately considering the anastomotic dehiscence as independent from the subsequent ARDS, renal failure, and eventually death. The Houston’s author only recorded the most severe complication (grade V, death) ignoring the other events due to the subjectivity in correlating one with the other. All other groups, however, agreed that complications of different nature must be recorded. Scenario 4, which describes a case with a poor general condition at the time of discharge, but without any specific complication, was labeled as having a “temporary disability” by the author from Houston, arguing that such results should be recorded, separately. All others failed to include in the classification a poor general condition at the time of discharge in the absence of specific complications. © 2009 Lippincott Williams & Wilkins
Annals of Surgery • Volume 250, Number 2, August 2009
TABLE 1. Center Scenario Case Case Case Case Case Case Case Case Case Case Case
1 2 3 4 5 6 7 8 9 10 11
Grading of Complications
Interpretation of the Scenarios Tokyo Japan (Makuuchi)
Buenos Aires Argentine (Santibañes/Pekolj)
Houston, USA (Vauthey)
Adelaide Australia (Padbury)
Baltimore USA (Schulick/Cameron)
Verona Italy (Bassi)
Montreal Canada (Barkun)
Zurich Switzerland (Clavien)
IIIa, IIIa V I, II, V No disability V V V No complication II IIIb IIIb
IIIa, IIIa V I, II, IIIb, V No disability V V V No complication II IIIb IIIb
IIIa, IIIa V V Disability V No complication V No complication II IIIb No complication
IIIa, IIIa V I, II, V No disability V V V No complication II IIIb IIIb
IIIa, IIIa V I, II, IIIb, V No disability V No complication V No complication II IIIb No complication
IIIa, IIIa V I, II, V No disability V V V No complication No complication IIIb No complication
IIIa, IIIa V I, II, V No disability V V V No complication II IIIb IIIb
IIIa, IIIa V I, II, V No disability V V V No complication II IIIb IIIb
tion. The results of this evaluation were discussed among the authors, and the consensus will be presented in the discussion.
Fourth Part: Correlation Between Severity of Each Respective Grade of Postoperative Complication and Perception Among Patients, Nurses, and Physicians
FIGURE 2. Perception of severity of postoperative complications related to the grade of complications by patients (䡺), nurses ( ), and physicians (f). Asterisk indicate significant difference as determined by nonparametric Kruskal-Wallis test (P ⬍ 0.05). Scenario 6 (death of the time of induction for surgery) was not considered a grade V complication by the Houston and Baltimore groups as death occurred before the incision. All other groups considered that death occurring in the operating room, even before incision, should be recorded. In scenario 9 (peptic ulcer after an inguinal hernia), the center form Verona, Italy suggested to ignore peptic ulcer after a minor surgery as a surgical complication. All other groups labeled it the same way as a grade II complication. In scenario 11 (incarcerated hernia after thyroid surgery), 3 centers (Houston and Baltimore and Verona, Italy) suggested to ignore the incarcerated hernia, as it was “obviously” unrelated to the first surgery. The 4 other centers recorded the complication as a grade IIIb complication. Other comments were provided. The group from Argentina underlined the need to record “bile duct injury” in scenario number 5 (death during laparoscopic surgery for acute cholecystitis). The same group also proposed a scenario, in which a patient developed excruciating pain during the early postoperative days, after a lung resection by lateral thoracotomy. As no underlying cause for the pain was identified, and the pain eventually disappeared on morphine therapy, the authors recorded this event as sequelae, but not as being a complication. All other groups agreed with this interpreta© 2009 Lippincott Williams & Wilkins
Thirty frequent and relevant postoperative course scenarios, each describing a specific complication, were independently graded by physicians, nurses, and patients (n ⫽ 50 in each group), using a numerical visual analogue scale from 0 (not severe) to 100 (extremely severe). The nonparametric test for multivariate analysis of variance (Kruskal-Wallis test) adapted with the Bonferroni correction, showed a significantly higher perception of surgical complications across all groups (grades I–IVb) by patients, when compared with physicians (P ⬍ 0.001) (Fig. 2). Patients perceived the severity of surgical complications significantly higher in grades IIIb, IVa, and IVb complications (grade IIIb: P ⬍ 0.001, 95% CI: ⫺11.36 to ⫺2.78; grade IVa: P ⬍ 0.01, 95% CI: ⫺9.34 to ⫺0.76, and grade IVb: P ⫽ 0.003, 95% CI: ⫺6.49 to ⫺2.09, respectively) than nurses. Nurses perceived the severity of grades II, IVa, and IVb complications significantly higher, when compared with physicians (grade II: P ⬍ 0.001, 95% CI: ⫺9.92 to ⫺1.53; grade IVa: P ⬍ 0.001, 95% CI: ⫺12.76 to ⫺4.36, and grade IVb: P ⬍ 0.001, 95% CI: ⫺7.69 to ⫺0.71, respectively).
DISCUSSION A transparent and ongoing assessment of quality in medicine is now required by patients, health care providers, and payers in many countries. However, any evaluation of surgical performance remains elusive, unless there is a common methodology for reporting negative outcomes.1 Reliable outcome data are crucial to improve surgical performance and gather credible data for benchmarking. Some countries report performance data available to patients to help them identify the “best” hospital or physician. Such exercises may potentially yield great danger in the absence of reliable markers of outcome. Another issue is the decreasing morbidity rates over recent years after many procedures,118,130,131 and, in those cases, differences in small numbers yield the risk of a high variability of data. Any system reporting complications must narrow the room for mistakes and subjective interpretation. Therefore, the challenge is to use a scale system, which, while simple, must not impede accuracy or general applicability. Many studies provide a measure of quality control by looking at M&M while predominantly listing individual complications without indication about the severity of the events. Attempts have been made in some areas to define and grade www.annalsofsurgery.com | 191
Clavien et al
frequent complications after specific procedures, such as fistula, delayed gastric emptying, and bleeding after pancreas surgery.132–134. Those systems are restricted to one type of surgery performed, and an across-the-board assessment of all complications is less likely achieved. Some groups have proposed systems of classifying complications in their own fields, such as colorectal surgeons from the Netherlands5 and general M&M surgical conference-based system from Boston.135 In 1992, one of us2 proposed 3 definitions for different types of negative outcomes after surgery (failure to cure, sequelae, and complication), completed by a system ranking complications according to severity.2 However, despite its use in a number of studies and large scale database,2,86,136 it failed to gain wide acceptance in literature. Therefore, we felt that some adjustments were necessary. With the participation of 10 centers around the world, we presented revisions in the grading system, proposing more objective criteria and some flexibility by allowing a contraction of grades to adjust to the patient population studied.7 In contrast to the initial classification,2 this revised system7 rapidly gained acceptance by a number of investigators, and became increasingly used in large scale trials. We, therefore, felt the need to perform a critical appraisal of the classification to enable either further endorsement or to highlight a need for further revision. A classification to be accepted by authors and readers must be understandable, simple, reproducible, and flexible. These criteria constituted the framework of the revised classification system.7 Before this publication, the grading system was evaluated by 150 surgeons from 10 centers around the world, indicating that the new proposal was indeed understandable, simple, and reproducible. For example, a 90% “correct answer rate” to rank examples of complications was documented among centers and surgeons at different levels of training. The classification was, however, not tested for complex complication scenarios and whether the grading system correlates with the perception by patients or health care providers. To test the general applicability of this classification, difficult case scenarios were collected over the past 5 years at the University of Zurich during the weekly M&M conferences. These scenarios were sent to other centers familiar with the classification system, asking them to grade each complication presented in the case scenarios. A high and highly significant degree of agreement (⬎90%) was documented, demonstrating that the classification also withstands in complex complication scenarios, retaining its reproducibility and objectivity. Discrepancies were noted only in a few scenarios. The most significant discussion focused on whether all complications must be graded separately, irrespective of their possible natural progression from one to the other, or whether only the most severe complication should be recorded (scenario 3). Four centers suggested recording only the most severe complication, when those of lesser severity are clearly a step in the process leading to the serious event. One center suggested registering only the most severe event, regardless of the nature of the other complications of lesser severity, while 2 centers separately recorded all complications. This scenario was again discussed among all authors, and a consensus was reached to label only the most severe events, when they are clearly related to each other; for example: a sequence of anastomotic dehiscence, sepsis, ARDS, renal failure, and death are recorded as grade V (death) attributable to the anastomotic breakdown. Another scenario, in which a postoperative complication occurred after an unrelated procedure (hernia after a thyroidectomy), triggered a dissent from 2 authors, who suggested ignoring the postoperative event. All others agreed to record all complications, irrespective of the likelihood of a link between both events, to prevent subjective interpretation and a risk of underreporting negative events. This was also the final 192 | www.annalsofsurgery.com
Annals of Surgery • Volume 250, Number 2, August 2009
consensus of all authors. Finally, the timing from which a complication should be recorded was discussed. Scenario 6 describing a death at the time of induction, that is, before skin incision was ignored by 2 groups. The authors reached the consensus to record as complication all events occurring in the operating room from the time of preparation for anesthesia. Finally, 1 group failed to record a peptic ulcer after a hernia repair due to uncertainty in the relationship between a minor procedure and a remote complication. The final consensus was to record all events to prevent subjective underreporting of negative events. Terms such as major, “moderate,” and minor are the main causes for confusion in outcome reporting. There are almost as many definitions for those terms as the number of investigators. This has not only led to misleading results, but also opened the door to “manipulation” of data. Using the grades of complications has provided more objective criteria for differentiating what may constitute a minor versus major complication, but here again the cut-off in the grading system has varied among authors. As in the original articles,2,7 we would strongly argue, to omit any such terminology, and to refer only to the specific grading. Those grades are sufficiently self-explanatory for patients and health care providers, and they do not need any additional labeling. Another source of confusion in the literature was the terminology given to the classification system, and how it was referred to, probably because no special denomination was provided in our original publication.7 Authors would like to refer to a name to prevent confusion with other systems. This classification is a revision of the inaugural report from 1992,2 which possibly explains that most authors referred to it as the “Clavien,” or “revised Clavien” classification or system, while about half of the authors did not provide any specific names. We would suggest recording and referencing this classification as the “Clavien-Dindo” classification. Some procedures are associated with few severe complications. The concept of the classification offers the flexibility of contracting or expanding the number of grades. Grade III covers 2 subgroups of complications requiring an invasive procedure, respectively, with (IIIa) and without (IIIb) the need of general anesthesia. Grade IV refers to organ failure with also 2 subdivisions for single (IVa) versus multiple (IVb) organ failures. We have used both the complete and contracted forms, depending on the need. In the literature, the classification has been used in the full forms in half of all reports. This purpose seems to have been well fulfilled. Of note, a few authors22,66,86,114,122 have adapted the initial classification system proposed in 1992,2 including features of the new classification7 to analyze specific procedures or to implement a “personalized” system in their institution. For example, living donor liver86 and kidney122 transplantation studies applied the “so-called” Clavien classification, but they based the system mostly on the initial classification,2,136 arguing that it was appropriate for their purpose, and that the prospective database was initiated before the introduction of the new system.7 This has resulted in a letter to the editor suggesting to respect the new system to prevent confusion and allow meaningful comparison between the data in Japan and United States.137 Another author from the field of urology122 proposed reporting surgical complications on the basis of criteria established at Memorial Sloan Kettering Cancer Center, New York, and United States.119 However, the Memorial Sloan Kettering Cancer Center criteria and grading system is similar to the initial classification.138,139 The current study provided further evidence from 3 different perspectives (patients, nurses, and physicians) of the staged severity of the grading by testing scenarios, illustrating a specific grade in the scale. Significant differences were noted among all grades including the 2 subtypes IIIa versus IIIb and IVa versus IVb, thereby also © 2009 Lippincott Williams & Wilkins
Annals of Surgery • Volume 250, Number 2, August 2009
validating differences among those subtypes of complications. This analysis also revealed new findings that may lead to further insight into the perception of various complications by physicians, nurses, and patients. For example, it revealed that patients perceive grades III and IV complications as more severe than nurses and physicians. A similar analysis revealed that women rank grades II and III complications at a significantly lower scale than men (data not shown). In conclusion, the current analysis demonstrates that the classification of surgical complications has reached its goal, and can be recommended in its current form for use in retrospective and prospective studies. This finding is important, as even with minor changes, a modified system will lead to confusion among authors and studies, at a time when many studies are under way. Indeed, large-scale trials will soon be published using this scale, which should lead to enhanced insight into the objective outcome after many surgical procedures. The examples presented in various scenarios should help users to grade complications in some difficult cases, and can be used for training by centers. Finally, to prevent the abuse of poorly defined terms, we recommend omitting any terms, such as minor, moderate, or major, as those terms will never be used consistently among authors. ACKNOWLEDGMENT The authors thank Milo A. Puhan, MD, PhD from Johns Hopkins University, and University Hospital Zurich, Horten Center, for his assistance in the statistic analysis. REFERENCES 1. Horton R. Surgical research or comic opera: questions, but few answers. Lancet. 1996;347:984 –985. 2. Clavien PA, Sanabria JR, Strasberg SM. Proposed classification of complications of surgery with examples of utility in cholecystectomy. Surgery. 1992;111:518 –526. 3. Pomposelli JJ, Gupta SK, Zacharoulis DC, et al. Surgical complication outcome (SCOUT) score: a new method to evaluate quality of care in vascular surgery. J Vasc Surg. 1997;25:1007–1014; discussion 1014 –1015. 4. Gawande AA, Thomas EJ, Zinner MJ, et al. The incidence and nature of surgical adverse events in Colorado and Utah in 1992. Surgery. 1999;126: 66 –75. 5. Veen MR, Lardenoye JW, Kastelein GW, et al. Recording and classification of complications in a surgical practice. Eur J Surg. 1999;165:421– 424; discussion 425. 6. Pillai SB, van Rij AM, Williams S, et al. Complexity-and risk-adjusted model for measuring surgical outcome. Br J Surg. 1999;86:1567–1572. 7. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240:205–213. 8. Chun YS, Vauthey JN, Ribero D, et al. Systemic chemotherapy and two-stage hepatectomy for extensive bilateral colorectal liver metastases: perioperative safety and survival. J Gastrointest Surg. 2007;11:1498 –1504; discussion 1504 –1505. 9. Ribero D, Abdalla EK, Madoff DC, et al. Portal vein embolization before major hepatectomy and its effects on regeneration, resectability and outcome. Br J Surg. 2007;94:1386 –1394. 10. Reddy SK, Morse MA, Hurwitz HI, et al. Addition of bevacizumab to irinotecan-and oxaliplatin-based preoperative chemotherapy regimens does not increase morbidity after resection of colorectal liver metastases. J Am Coll Surg. 2008;206:96 –106. 11. Reddy SK, Pawlik TM, Zorzi D, et al. Simultaneous resections of colorectal cancer and synchronous liver metastases: a multi-institutional analysis. Ann Surg Oncol. 2007;14:3481–3491. 12. de Santibanes E, Ardiles V, Gadano A, et al. Liver transplantation: the last measure in the treatment of bile duct injuries. World J Surg. 2008;32:1714 – 1721. 13. Seda-Neto J, Godoy AL, Carone E, et al. Left lateral segmentectomy for pediatric live-donor liver transplantation: special attention to segment IV complications. Transplantation. 2008;86:697–701. 14. Sundaram CP, Martin GL, Guise A, et al. Complications after a 5-year
© 2009 Lippincott Williams & Wilkins
Grading of Complications
15.
16.
17.
18. 19.
20.
21.
22.
23. 24. 25.
26. 27.
28.
29.
30.
31.
32. 33.
34.
35.
36.
37.
38.
39.
experience with laparoscopic donor nephrectomy: the Indiana University experience. Surg Endosc. 2007;21:724 –728. McKay A, Sutherland FR, Bathe OF, et al. Morbidity and mortality following multivisceral resections in complex hepatic and pancreatic surgery. J Gastrointest Surg. 2008;12:86 –90. Tamura S, Sugawara Y, Kaneko J, et al. Systematic grading of surgical complications in live liver donors according to Clavien’s system. Transpl Int. 2006;19:982–987. Pessaux P, Regimbeau JM, Dondero F, et al. Randomized clinical trial evaluating the need for routine nasogastric decompression after elective hepatic resection. Br J Surg. 2007;94:297–303. Permpongkosol S, Link RE, Su LM, et al. Complications of 2775 urological laparoscopic procedures: 1993 to 2005. J Urol. 2007;177:580 –585. Patel S, Cassuto J, Orloff M, et al. Minimizing morbidity of organ donation: analysis of factors for perioperative complications after living-donor nephrectomy in the United States. Transplantation. 2008;85:561–565. Reddy S, Edil BH, Cameron JL, et al. Pancreatic resection of isolated metastases from nonpancreatic primary cancers. Ann Surg Oncol. 2008;15: 3199 –3206. Haynes AB, Weiser TG, Berry WR, et al. A surgical safety checklist to reduce morbidity and mortality in a global population. N Engl J Med. 2009;360:491– 499. Barr ML, Belghiti J, Villamil FG, et al. A report of the Vancouver Forum on the care of the live organ donor: lung, liver, pancreas, and intestine data and medical guidelines. Transplantation. 2006;81:1373–1385. Fan ST. Live donor liver transplantation in adults. Transplantation. 2006; 82:723–732. Sugawara Y, Tamura S, Makuuchi M. Systematic grading of surgical complications in live liver donors. Liver Transpl. 2007;13:781–782. Tamura S, Sugawara Y, Kokudo N. Donor evaluation and hepatectomy for living-donor liver transplantation. J Hepatobiliary Pancreat Surg. 2008;15: 79 –91. Skolarikos A, de la Rosette J. Prevention and treatment of complications following percutaneous nephrolithotomy. Curr Opin Urol. 2008;18:229–234. Breitenstein S, Kraus A, Hahnloser D, et al. Emergency left colon resection for acute perforation: primary anastomosis or Hartmann’s procedure? A case-matched control study. World J Surg. 2007;31:2117–2124. Leandros E, Alexakis N, Albanopoulos K, et al. Hand-assisted laparoscopic surgery with a Pfannenstiel incision in beta-thalassemia patients: initial experience. World J Surg. 2006;30:1216 –1220. Hubner M, Hahnloser D, Hetzer F, et al. A prospective randomized comparison of two instruments for dissection and vessel sealing in laparoscopic colorectal surgery. Surg Endosc. 2007;21:592–594. Scholz T, Hetzer FH, Dindo D, et al. Long-term follow-up after combined fissurectomy and botox injection for chronic anal fissures. Int J Colorectal Dis. 2007;22:1077–1081. Kyoden Y, Tamura S, Sugawara Y, et al. Biliary complications in right lateral sector graft live donor liver transplantation. Transpl Int. 2008;21: 332–339. Bingener J, Buck L, Richards M, et al. Long-term outcomes in laparoscopic vs. open ventral hernia repair. Arch Surg. 2007;142:562–567. Arru M, Pulitano C, Aldrighetti L, et al. A prospective evaluation of ultrasonic dissector plus harmonic scalpel in liver resection. Am Surg. 2007;73:256 –260. Jaffe J, Stakhovsky O, Cathelineau X, et al. Surgical outcomes for men undergoing laparoscopic radical prostatectomy after transurethral resection of the prostate. J Urol. 2007;178:483– 487; discussion 487. Zorzi D, Chun YS, Madoff DC, et al. Chemotherapy with bevacizumab does not affect liver regeneration after portal vein embolization in the treatment of colorectal liver metastases. Ann Surg Oncol. 2008;15:2765–2772. Mullen JT, Ribero D, Reddy SK, et al. Hepatic insufficiency and mortality in 1059 noncirrhotic patients undergoing major hepatectomy. J Am Coll Surg. 2007;204:854 – 862; discussion 862– 864. Petrowsky H, McCormack L, Trujillo M, et al. A prospective, randomized, controlled trial comparing intermittent portal triad clamping versus ischemic preconditioning with continuous clamping for major liver resection. Ann Surg. 2006;244:921–928; discussion 928 –930. Lesurtel M, Selzner M, Petrowsky H, et al. How should transection of the liver be performed? A prospective randomized study in 100 consecutive patients: comparing four different transection strategies. Ann Surg. 2005; 242:814 – 822; discussion 822– 823. Wicherts DA, Miller R, de Haas RJ, et al. Long-term results of two-stage hepatectomy for irresectable colorectal cancer liver metastases. Ann Surg.
www.annalsofsurgery.com | 193
Clavien et al
2008;248:994 –1005. 40. Ko EY, Castle EP, Desai PJ, et al. Utility of the endovascular stapler for right-sided laparoscopic donor nephrectomy: a 7-year experience at Mayo Clinic. J Am Coll Surg. 2008;207:896 –903. 41. Heinrich S, Schafer M, Weber A, et al. Neoadjuvant chemotherapy generates a significant tumor response in resectable pancreatic cancer without increasing morbidity: results of a prospective phase II trial. Ann Surg. 2008;248:1014 –1022. 42. de Rougemont O, Dutkowski P, Weber M, et al. Abdominal drains in liver transplantation: useful tool or useless dogma? A matched case-control study. Liver Transpl. 2009;15:96 –101. 43. Constantinides CA, Tyritzis SI, Skolarikos A, et al. Short- and long-term complications of open radical prostatectomy according to the Clavien classification system. BJU Int. 2009;103:336 –340. 44. Yoon YS, Han HS, Shin SH, et al. Laparoscopic treatment for intrahepatic duct stones in the era of laparoscopy: laparoscopic intrahepatic duct exploration and laparoscopic hepatectomy. Ann Surg. 2009;249:286 –291. 45. Shen P, Thai K, Stewart JH, et al. Peritoneal surface disease from colorectal cancer: comparison with the hepatic metastases surgical paradigm in optimally resected patients. Ann Surg Oncol. 2008;15:3422–3432. 46. de la Rosette JJ, Zuazu JR, Tsakiris P, et al. Prognostic factors and percutaneous nephrolithotomy morbidity: a multivariate analysis of a contemporary series using the Clavien classification. J Urol. 2008;180:2489 – 2493. 47. Muller MK, Dedes KJ, Dindo D, et al. Impact of clinical pathways in surgery. Langenbecks Arch Surg. 2009;394:31–39. 48. Fuks D, Verhaeghe P, Brehant O, et al. Results of laparoscopic sleeve gastrectomy: a prospective study in 135 patients with morbid obesity. Surgery. 2009;145:106 –113. 49. Bonvalot S, Rivoire M, Castaing M, et al. Primary retroperitoneal sarcomas: a multivariate analysis of surgical factors associated with local control. J Clin Oncol. 2009;27:31–37. 50. Gumbs AA, Gres P, Madureira FA, et al. Laparoscopic vs. open resection of noninvasive intraductal pancreatic mucinous neoplasms. J Gastrointest Surg. 2008;12:707–712. 51. Gumbs AA, Gres P, Madureira F, et al. Laparoscopic vs. open resection of pancreatic endocrine neoplasms: single institution’s experience over 14 years. Langenbecks Arch Surg. 2008;393:391–395. 52. Beck-Schimmer B, Breitenstein S, Urech S, et al. A randomized controlled trial on pharmacological preconditioning in liver surgery using a volatile anesthetic. Ann Surg. 2008;248:909 –918. 53. Erekson EA, Sung VW, Myers DL. Pay for performance: what the urogynecologist should know. Int Urogynecol J Pelvic Floor Dysfunct. 2008;19: 1039 –1041. 54. Chan SC, Fan ST. Historical perspective of living donor liver transplantation. World J Gastroenterol. 2008;14:15–21. 55. Gonzalgo ML, Pavlovich CP, Trock BJ, et al. Classification and trends of perioperative morbidities following laparoscopic radical prostatectomy. J Urol. 2005;174:135–139; discussion 139. 56. Yi NJ, Suh KS, Cho JY, et al. Three-quarters of right liver donors experienced postoperative complications. Liver Transpl. 2007;13:797– 806. 57. Lermite E, Pessaux P, Teyssedou C, et al. Effect of preoperative endoscopic biliary drainage on infectious morbidity after pancreatoduodenectomy: a case-control study. Am J Surg. 2008;195:442– 446. 58. Patel S, Orloff M, Tsoulfas G, et al. Living-donor liver transplantation in the United States: identifying donors at risk for perioperative complications. Am J Transplant. 2007;7:2344 –2349. 59. Bucher P, Oulhaci W, Morel P, et al. Results of conservative treatment for perforated gastroduodenal ulcers in patients not eligible for surgical repair. Swiss Med Wkly. 2007;137:337–340. 60. Lucidi V, Lemye AC, Baire L, et al. Use of marginal donors for liver transplantation: a single-center experience within the Eurotransplant patientdriven allocation system. Transplant Proc. 2007;39:2668 –2671. 61. Pratt W, Joseph S, Callery MP, et al. POSSUM accurately predicts morbidity for pancreatic resection. Surgery. 2008;143:8 –19. 62. Feldman LS, Demyttenaere SV, Polyhronopoulos GN, et al. Refining the selection criteria for laparoscopic versus open splenectomy for splenomegaly. J Laparoendosc Adv Surg Tech A. 2008;18:13–19. 63. Lermite E, Pessaux P, Brehant O, et al. Risk factors of pancreatic fistula and delayed gastric emptying after pancreaticoduodenectomy with pancreaticogastrostomy. J Am Coll Surg. 2007;204:588 –596. 64. Stolzenburg JU, Rabenalt R, Do M, et al. Categorisation of complications of endoscopic extraperitoneal and laparoscopic transperitoneal radical prosta-
194 | www.annalsofsurgery.com
Annals of Surgery • Volume 250, Number 2, August 2009
tectomy. World J Urol. 2006;24:88 –93. 65. Tan HP, Shapiro R, Montgomery RA, et al. Proposed live donor nephrectomy complication classification scheme. Transplantation. 2006;81:1221– 1223. 66. Kocak B, Koffron AJ, Baker TB, et al. Proposed classification of complications after live donor nephrectomy. Urology. 2006;67:927–931. 67. Goeman L, Salomon L, La De Taille A, et al. Long-term functional and oncological results after retroperitoneal laparoscopic prostatectomy according to a prospective evaluation of 550 patients. World J Urol. 2006;24:281–288. 68. Teber D, Tefekli A, Eskicorapci S, et al. Retroperitoneoscopy: a Versatile Access for Many Urologic Indications. Eur Urol Suppl 2006;5:975–982. 69. Chan SC, Fan ST, Lo CM, et al. Toward current standards of donor right hepatectomy for adult-to-adult live donor liver transplantation through the experience of 200 cases. Ann Surg. 2007;245:110 –117. 70. Vanounou T, Pratt W, Fischer JE, et al. Deviation-based cost modeling: a novel model to evaluate the clinical and economic impact of clinical pathways. J Am Coll Surg. 2007;204:570 –579. 71. Vanounou T, Pratt WB, Callery MP, et al. Selective administration of prophylactic octreotide during pancreaticoduodenectomy: a clinical and cost-benefit analysis in low- and high-risk glands. J Am Coll Surg. 2007; 205:546 –557. 72. Vollmer CM Jr, Pratt W, Vanounou T, et al. Quality assessment in highacuity surgery: volume and mortality are not enough. Arch Surg. 2007;142: 371–380. 73. Bergman S, Feifer A, Feldman LS, et al. Laparoscopic live donor nephrectomy: the pediatric recipient in a dual-site program. Pediatr Transplant. 2007;11:429 – 432. 74. Gali B, Findlay JY, Plevak DJ, et al. Right hepatectomy for living liver donation vs. right hepatectomy for disease: intraoperative and immediate postoperative comparison. Arch Surg. 2007;142:467–471; discussion 471–472. 75. Khalaf H, Al-Sofayan M, El-Sheikh Y, et al. Donor outcome after living liver donation: a single-center experience. Transplant Proc. 2007;39:829–834. 76. Yi NJ, Suh KS, Lee HW, et al. An artificial vascular graft is a useful interpositional material for drainage of the right anterior section in living donor liver transplantation. Liver Transpl. 2007;13:1159 –1167. 77. Tefekli A, Ali Karadag M, Tepeler K, et al. Classification of percutaneous nephrolithotomy complications using the modified Clavien grading system: looking for a standard. Eur Urol. 2008;53:184 –190. 78. Bagry H, de la Cuadra Fontaine JC, Asenjo JF, et al. Effect of a continuous peripheral nerve block on the inflammatory response in knee arthroplasty. Reg Anesth Pain Med. 2008;33:17–23. 79. Zavorsky GS, Kim do J, Sylvestre JL, et al. Alveolar-membrane diffusing capacity improves in the morbidly obese after bariatric surgery. Obes Surg. 2008;18:256 –263. 80. Bergman S, Feldman LS, Anidjar M, et al. “First, do no harm”: monitoring outcomes during the transition from open to laparoscopic live donor nephrectomy in a Canadian centre. Can J Surg. 2008;51:103–110. 81. Schieman C, MacLean AR, Buie WD, et al. Does surgeon fatigue influence outcomes after anterior resection for rectal cancer? Am J Surg. 2008;195: 684 – 687; discussion 687– 688. 82. Finley DS, Beck S, Box G, et al. Percutaneous and laparoscopic cryoablation of small renal masses. J Urol. 2008;180:492– 498; discussion 498. 83. Regenbogen SE, Lancaster RT, Lipsitz SR, et al. Does the Surgical Apgar Score measure intraoperative performance? Ann Surg. 2008;248:320 –328. 84. Yi NJ, Suh KS, Cho YB, et al. The right small-for-size graft results in better outcomes than the left small-for-size graft in adult-to-adult living donor liver transplantation. World J Surg. 2008;32:1722–1730. 85. Ishizaki Y, Yoshimoto J, Sugo H, et al. Hepatectomy using traditional Pean clamp-crushing technique under intermittent Pringle maneuver. Am J Surg. 2008;196:353–357. 86. Ghobrial RM, Freise CE, Trotter JF, et al. Donor morbidity after living donation for liver transplantation. Gastroenterology. 2008;135:468 – 476. 87. Bellavance EC, Lumpkins KM, Mentha G, et al. Surgical management of early-stage hepatocellular carcinoma: resection or transplantation? J Gastrointest Surg. 2008;12:1699 –1708. 88. Kondo K, Chijiiwa K, Funagayama M, et al. Hepatic resection is justified for elderly patients with hepatocellular carcinoma. World J Surg. 2008;32: 2223–2229. 89. Medina-Franco H, Garcia-Alvarez MN, Ortiz-Lopez LJ, et al. Predictors of adverse surgical outcome in the management of malignant bowel obstruction. Rev Invest Clin. 2008;60:212–216.
© 2009 Lippincott Williams & Wilkins
Annals of Surgery • Volume 250, Number 2, August 2009
90. Gjertson CK, Sundaram CP. Testicular pain following laparoscopic renal surgery. J Urol. 2008;180:2037–2040; discussion 2040 –2041. 91. Matsui Y, Sugawara Y, Yamashiki N, et al. Living donor liver transplantation for fulminant hepatic failure. Hepatol Res. 2008;38:987–996. 92. Goh BK, Tan YM, Chung YF, et al. Critical appraisal of 232 consecutive distal pancreatectomies with emphasis on risk factors, outcome, and management of the postoperative pancreatic fistula: a 21-year experience at a single institution. Arch Surg. 2008;143:956 –965. 93. de Haas RJ, Wicherts DA, Flores E, et al. R1 resection by necessity for colorectal liver metastases: is it still a contraindication to surgery? Ann Surg. 2008;248:626 – 637. 94. Friedman AL, Peters TG, Jones KW, et al. Fatal and nonfatal hemorrhagic complications of living kidney donation. Ann Surg. 2006;243:126 –130. 95. Porpiglia F, Terrone C, Tarabuzzi R, et al. Transperitoneal versus extraperitoneal laparoscopic radical prostatectomy: experience of a single center. Urology. 2006;68:376 –380. 96. Albo ME, Richter HE, Brubaker L, et al. Burch colposuspension versus fascial sling to reduce urinary stress incontinence. N Engl J Med. 2007;356: 2143–2155. 97. Zullo MA, Plotti F, Calcagno M, et al. One-year follow-up of tension-free vaginal tape (TVT) and trans-obturator suburethral tape from inside to outside (TVT-O) for surgical treatment of female stress urinary incontinence: a prospective randomized trial. Eur Urol. 2007;51:1376 –1382; discussion 1383–1384. 98. Laurent C, Leblanc F, Gineste C, et al. Laparoscopic approach in surgical treatment of rectal cancer. Br J Surg. 2007;94:1555–1561. 99. Laurent C, Leblanc F, Bretagnol F, et al. Long-term wound advantages of the laparoscopic approach in rectal cancer. Br J Surg. 2008;95:903–908. 100. Zullo MA, Ruggiero A, Plotti F, et al. Anterior colporrhaphy plus inside-out tension-free vaginal tape for associated stress urinary incontinence and cystocele. J Minim Invasive Gynecol. 2008;15:446 – 451. 101. Alperin M, Sutkin G, Ellison R, et al. Perioperative outcomes of the Prolift(R) pelvic floor repair systems following introduction to a urogynecology teaching service. Int Urogynecol J Pelvic Floor Dysfunct. 2008;19: 1617–1622. 102. Lelong B, Bege T, Esterni B, et al. Short-term outcome after laparoscopic or open restorative mesorectal excision for rectal cancer: a comparative cohort study. Dis Colon Rectum. 2007;50:176 –183. 103. Strasberg SM, Linehan DC, Clavien PA, et al. Proposal for definition and severity grading of pancreatic anastomosis failure and pancreatic occlusion failure. Surgery. 2007;141:420 – 426. 104. Rozet F, Jaffe J, Braud G, et al. A direct comparison of robotic assisted versus pure laparoscopic radical prostatectomy: a single institution experience. J Urol. 2007;178:478 – 482. 105. Strasberg SM, Linehan DC, Hawkins WG. Radical antegrade modular pancreatosplenectomy procedure for adenocarcinoma of the body and tail of the pancreas: ability to obtain negative tangential margins. J Am Coll Surg. 2007;204:244 –249. 106. Nadalin S, Malago M, Radtke A, et al. Current trends in live liver donation. Transpl Int. 2007;20:312–330. 107. Belli G, Fantini C, D’Agostino A, et al. Laparoscopic versus open liver resection for hepatocellular carcinoma in patients with histologically proven cirrhosis: short- and middle-term results. Surg Endosc. 2007;21:2004 –2011. 108. Elias D, Goere D, Di Pietrantonio D, et al. Results of systematic second-look surgery in patients at high risk of developing colorectal peritoneal carcinomatosis. Ann Surg. 2008;247:445– 450. 109. Mohebati A, Schwarz RE. Extended left-sided pancreatectomy with spleen preservation. J Surg Oncol. 2008;97:150 –155. 110. Murakami Y, Uemura K, Hayashidani Y, et al. No mortality after 150 consecutive pancreatoduodenctomies with duct-to-mucosa pancreaticogastrostomy. J Surg Oncol. 2008;97:205–209. 111. Elias D, Honore C, Ciuchendea R, et al. Peritoneal pseudomyxoma: results of a systematic policy of complete cytoreductive surgery and hyperthermic intraperitoneal chemotherapy. Br J Surg. 2008;95:1164 –1171. 112. Aldrighetti L, Pulitano C, Catena M, et al. A prospective evaluation of laparoscopic versus open left lateral hepatic sectionectomy. J Gastrointest Surg. 2008;12:457– 462. 113. Muscari F, Suc B, Kirzin S, et al. Risk factors for mortality and intraabdominal complications after pancreatoduodenectomy: multivariate analysis in 300 patients. Surgery. 2006;139:591–598.
© 2009 Lippincott Williams & Wilkins
Grading of Complications
114. Veen EJ, Steenbruggen J, Roukema JA. Classifying surgical complications: a critical appraisal. Arch Surg. 2005;140:1078 –1083. 115. Bargman V, Sundaram CP, Bernie J, et al. Randomized trial of laparoscopic donor nephrectomy with and without hand assistance. J Endourol. 2006;20: 717–722. 116. Noblett SE, Snowden CP, Shenton BK, et al. Randomized clinical trial assessing the effect of Doppler-optimized fluid management on outcome after elective colorectal resection. Br J Surg. 2006;93:1069 –1076. 117. Hwang S, Lee SG, Lee YJ, et al. Lessons learned from 1000 living donor liver transplantations in a single center: how to make living donations safe. Liver Transpl. 2006;12:920 –927. 118. DeOliveira ML, Winter JM, Schafer M, et al. Assessment of complications after pancreatic surgery: a novel grading system applied to 633 patients undergoing pancreaticoduodenectomy. Ann Surg.2006; 244:931–937; discussion 937–939. 119. Martin RC II, Brennan MF, Jaques DP. Quality of complication reporting in the surgical literature. Ann Surg. 2002;235:803– 813. 120. Guo DY, Eteuati J, Nguyen MH, et al. Laparoscopic assisted colectomy: experience from a rural centre. ANZ J Surg. 2007;77:283–286. 121. Wolff BG, Weese JL, Ludwig KA, et al. Postoperative ileus-related morbidity profile in patients treated with alvimopan after bowel resection. J Am Coll Surg. 2007;204:609 – 616. 122. Donat SM. Standards for surgical complication reporting in urologic oncology: time for a change. Urology. 2007;69:221–225. 123. de Reuver PR, Grossmann I, Busch OR, et al. Referral pattern and timing of repair are risk factors for complications after reconstructive surgery for bile duct injury. Ann Surg. 2007;245:763–770. 124. McKay A, Motamedi M, Temple W, et al. Vascular reconstruction with the superficial femoral vein following major oncologic resection. J Surg Oncol. 2007;96:151–159. 125. Nakano H, Oussoultzoglou E, Rosso E, et al. Sinusoidal injury increases morbidity after major hepatectomy in patients with colorectal liver metastases receiving preoperative chemotherapy. Ann Surg. 2008;247:118 –124. 126. Schiesser M, Chen JW, Maddern GJ, et al. Perioperative morbidity affects long-term survival in patients following liver resection for colorectal metastases. J Gastrointest Surg. 2008;12:1054 –1060. 127. McCormack L, Petrowsky H, Jochum W, et al. Hepatic steatosis is a risk factor for postoperative complications after major hepatectomy: a matched case-control study. Ann Surg. 2007;245:923–930. 128. Hubner M, Demartines N, Muller S, et al. Prospective randomized study of monopolar scissors, bipolar vessel sealer and ultrasonic shears in laparoscopic colorectal surgery. Br J Surg. 2008;95:1098 –1104. 129. Muller S, Zalunardo MP, Hubner M, et al. A fast-track program reduces complications and length of hospital stay after open colonic surgery. Gastroenterology. 2009;136:842– 847. 130. Birkmeyer JD, Stukel TA, Siewers AE, et al. Surgeon volume and operative mortality in the United States. N Engl J Med. 2003;349:2117–2127. 131. Birkmeyer JD, Siewers AE, Finlayson EV, et al. Hospital volume and surgical mortality in the United States. N Engl J Med. 2002;346:1128 –1137. 132. Bassi C, Dervenis C, Butturini G, et al. Postoperative pancreatic fistula: an international study group (ISGPF) definition. Surgery. 2005;138:8 –13. 133. Wente MN, Bassi C, Dervenis C, et al. Delayed gastric emptying (DGE) after pancreatic surgery: a suggested definition by the International Study Group of Pancreatic Surgery (ISGPS). Surgery. 2007;142:761–768. 134. Wente MN, Veit JA, Bassi C, et al. Postpancreatectomy hemorrhage (PPH): an International Study Group of Pancreatic Surgery (ISGPS) definition. Surgery. 2007;142:20 –25. 135. Antonacci AC, Lam S, Lavarias V, et al. A morbidity and mortality conference-based classification system for adverse events: surgical outcome analysis: part I. J Surg Res. 2008;147:172–177. 136. Clavien PA, Camargo CA Jr, Croxford R, et al. Definition and classification of negative outcomes in solid organ transplantation: application in liver transplantation. Ann Surg. 1994;220:109 –120. 137. Tamura S, Sugawara Y, Kukudo N, et al. Systematic grading of morbidity after living donation for liver transplantation. Gastroenterology. 2008;135:1804. 138. D’Angelica M, Kornprat P, Gonen M, et al. Lack of evidence for increased operative morbidity after hepatectomy with perioperative use of bevacizumab: a matched case-control study. Ann Surg Oncol. 2007;14:759 –765. 139. DeMatteo RP, Maki RG, Singer S, et al. Results of tyrosine kinase inhibitor therapy followed by surgical resection for metastatic gastrointestinal stromal tumor. Ann Surg. 2007;245:347–352.
www.annalsofsurgery.com | 195
Annals of Surgery • Volume 250, Number 2, August 2009
Clavien et al
APPENDIX A. Classification of Surgical Complications Grades Grade I:
Grade II: Grade III: Grade III-a: Grade III-b: Grade IV: Grade IV-a: Grade IV-b: Grade V: Suffix ’d’:
Definition Any deviation from the normal postoperative course without the need for pharmacological treatment or surgical, endoscopic and radiological interventions. Acceptable therapeutic regimens are: drugs as antiemetics, antipyretics, analgetics, diuretics and electrolytes and physiotherapy. This grade also includes wound infections opened at the bedside. Requiring pharmacological treatment with drugs other than such allowed for grade I complications. Blood transfusions and total parenteral nutrition are also included. Requiring surgical, endoscopic or radiological intervention intervention not under general anesthesia intervention under general anesthesia Life-threatening complication (including CNS complications)‡ requiring IC/ICU-management single organ dysfunction (including dialysis) multi organ dysfunction Death of a patient If the patient suffers from a complication at the time of discharge (see examples in Appendix B, http://Links.Lww-.com/SLA/A3), the suffix “d” (for ‘disability’) is added to the respective grade of complication. This label indicates the need for a follow-up to fully evaluate the complication.
‡ brain hemorrhage, ischemic stroke, subarrachnoidal bleeding, but excluding transient ischemic attacks (TIA); IC: Intermediate care; ICU: Intensive care unit www.surgicalcomplication.info
196 | www.annalsofsurgery.com
© 2009 Lippincott Williams & Wilkins
