Oral mannitol as a preparation for double contrast barium enema

ClinicalRadiology (1982) 33,467-469 © 1982 Royal College of Radiologists

0009-9260/82/00800467502.00

Oral Mannitol as a Preparation for Double Contrast Barium Enema K. D. FOORD

Department of Radiology, Royal Hallamshire Hospital, Glossop Road, Sheffield An oral mannitol purgative which is used prior to colonoscopy proved disappointing for barium enema preparation. A number of ways of improving bowel clearance with mannitol in divided dosage, preliminary oral metoclopramide and a clear fluid diet have been assessed. Divided dosage and a clear fluid diet were both beneficial, whereas oral metoclopramide was ineffective. Using the best regime, 38% of patients had a totally clear bowel compared with 33% using an established lavage technique. It is concluded that oral mannitol, with a clear fluid diet, is a satisfactory preparation for double contrast barium enema, and a regime is recommended.

Oral mannitol when taken in quantities greater than At 18.00 hours on the day prior to the examination 40g will consistently produce osmotic diarrhoea patients were instructed to drink 1 litre of 10% (Nasrullah and Iber, 1969), thus acting as an ante- mannitol solution as quickly as possible, after which grade bowel washout. Mannitol is frequently used they were encouraged to drink plenty of fluids but for bowel preparation prior to surgery and colono- have nothing to eat. scopy and has minimal metabolic effects (Minervini One hundred and eighty-three consecutive et al., 1980). examinations were performed but 33 studies were Palmer and Khan (1979) have shown that satis- excluded due to failure to comply precisely with factory preparation of the large bowel for barium the instructions. The remaining 150 examinations enema can be obtained with an oral mannitol tech- available for analysis were all performed at routine nique based on that used for colonoscopy. However, out-patient sessions, 36 in the morning after 11.00 results are inconsistent and excess fluid retention in hours, and 114 in the afternoon after 14.00 hours. the colon can have an adverse effect on mucosal One hundred consecutive patients within this group coating. This regime was modified to include a low were also asked to complete a questionnaire and 84 residue diet with the patient instructed to drink the o f these proved suitable for analysis. mannitol 12h earlier to allow the fluid to be reabsorbed. This modified regime (the 'standard' mannitol preparation) was reviewed after 1 year, and Part 2 Study: 'Standard' Oral Mannitol vs. Bisacodyl was additionally compared with other modifications + Lavage and colonic lavage. This review suggested that bowel At the Northern General Hospital, 27 out-patients clearance is improved by reducing the incidence of were prepared using the 'standard' mannitol techvomiting, decreasing post-diarrhoeal small bowel nique concurrently with a group of 18 age-matched spill to the caecum and increasing the duration of out-patients prepared with a combination of low diarrhoea. In order to do this, slower administration residue diet for 3 days, bisacodyl 15 mg orally for of mannitol, the use of an anti-emetic (metoclopra- two preceding evenings and colonic lavage by mide) and a clear fluid diet were tried and assessed. experienced nursing staff 2 h before the barium examination.

METHOD

Part 1 Study: 'Standard' Oral Mannitol Preparation

Part 3 Study: (A) 'Standard' Oral Mannitol vs. (B) Divided Dose Oral Mannitol vs. (C) Divided Dose Oral Mannitol + Metoclopramide vs. (D) Divided Dose Mannitol + Metoclopramide + Clear Fluid Diet

A low residue diet was prescribed f o r 3 days before double contrast barium enema examination.

Three modifications were compared with a control group prepared with the 'standard' mannitol regime.

The trial was condicted in three parts.

468

CLINICAL RADIOLOGY

Forty-five out-patients were allocated at random to each o f four preparation regimes: (A) the 'standard' regime; (B) as (A) b u t the patients received instructions to drink the mannitol solution in four divided 250 ml doses at 1 0 - 1 5 min intervals; (C) as (B) but with 20 mg metoclopramide in tablet form taken 3 0 m i n before t h e first drink o f mannitol; (D) as (C) but with the addition o f a clear fluid diet from breakfast on the day before the examination. The patients were randomly examined at either morning or afternoon sessions. F o r administrative reasons several patients were excluded, and the number remaining in each group was about 37. ANALYSIS O F M A T E R I A L The examinations were analysed as follows: All examinations were assessed b y two radiologists together and a grading system applied. Neither radiologist whilst grading was aware o f the preparation group o f any of the patients. Each examination was graded according to the following criteria: Excellent: A totally clear large bowel, with no trace of faeces. Good: Some particulate debris of less than 10 mm diameter allowable in the right hemicolon b u t the left hemicolon clear. Moderate: Scattered particulate matter present, such that polyps of less than 1 0 m m diameter could not be excluded. Poor: Considerable faecal residue such that only a gross lesion could be excluded.

RESULTS The results o f the three studies are shown in the three tables, and the following conclusions are made: Part 1 Study (Table 1) There is a trend to better colon cleansing by delaying the examination until the afternoon. An additional finding was that vomiting reduced clearance by on average one grade. In most cases diarrhoea commenced between 15 min and 3 h from drinking the first mannitol. There was a trend towards more prolonged diarrhoea and better bowel clearance if drinking time was extended. The diarrhoea lasted approximately 6 h with eight bowel actions (median figures).

Table 1 - Grading of double contrast barium enemas asia. the 'standard' mannitol preparation (Part 1 Study) Grade

Excellent Good Moderate Poor Totals

Morning

Afternoon

Number %

Number %

8 7 15 6 36

18 48 33 15 114

-

22.2 } 19.4 41.6 41.8 16.6

15.8 } 42.1 57.9 28.9 13.2

Applied to the whole table: x 2 = 6.09; n = 3; 0.15 > p ~ 0.1 (not significant). Grouping Moderate and Poor results;* ×2 = 6.04; n = 2; 0.05 > P > 0.025 (probably significant at the 5% level that afternoon results are better than morning results). *A unit association test confirmed that this was statistic. ally acceptable as the differences between these two groups were more subjective than the differences between Excellent, Good and Moderate groups. Table 2 - Comparative clearance grading between age. matched groups of patients prepared by the 'standard' mannitol technique and colonic lavage (Part 2 Study) Grade

Mannitol

Lavage

Number

%

Excellent Good Moderate Poor

5 10 7 } 5 12

18.5 37.1 25.9 18.5

Totals

27

Number

6 8 2 } 2 4

33.3 44.5 11.1 11.1

18

Comparing Excellent and Good results with grouped Moderate + Poor results: ×2 = 2.69; n = 2; 0.3 > P > 0.2 (no significant differences between the two techniques).

Part 2 Study (Table 2) Colonic lavage was not shown to be significantly better than the 'standard' mannitol regime. Part 3 Study (Table 3) The results o f the various modifications are compared with the 'standard' regime in Table 3. Regime (D) gave the best results, significantly better than the 'standard' regime (A). Using this regime nearly 40% o f patients had totally clear colons which were considered ideal for double-contrast barium studies. A further 15% had some residue in the right hemicolon and only one out o f 39 was so poor that only a gross lesion could be excluded. The incidence o f vomiting was reduced to just over 10% b y dividing the mannitol dosage, but remained at 25% for regime (D) which involved a fluid-only diet. Prior oral metoclopramide was of no benefit.

O R A L M A N N I T O L AS A P R E P A R A T I O N FOR DOUBLE C O N T R A S T BARIUM ENEMA Table 3 - Effect of trial mannitol regimes on grading of preparation (Part 3 Study) rrade

A

B

C

D

Nok %

No. %

No.

Excellent Good Moderate poor

4 14 14 5

9 12 11 6

7 21.2 11 33.3 1236.4 3 9.1

15 6 17 1

Totals

37

33

39

%

No. %

f

10.8 37.9 37.9 13.4

38

23.7 31.6 29.0 15.7

38.4 15.4 43.6 2.6

Applying X2 to the whole table after grouping Moderate + poor results: × z = 10.77; n = 6 ; 0.1 > P > 0.05 (on the borderline of statistical significance). Avs. B: Xz = 2.17;n B vs. C: X2 = 0.27; n Cvs. D: ×2 = 4.24;n Avs. D:X 2 = 9.55;n

= 2;0.4 > P > 0.3 = 2; 0.9 > P > 0.85 = 2;0.15 > P > 0 . 1 = 2;0.01 > P > 0 . 0 0 5

There is a significant superiority of regime D over regime A. There is a trend towards concordance between regime B and C. No other significant differences are shown, although on trend D seems better than C.

469

CONCLUSION As a result o f our studies it is c o n c l u d e d that the most effective m a n n i t o l regime is as follows: a low residue diet f r o m 3 days before the enema, w i t h clear fluids only f r o m after breakfast on the day before. No fluids for 2 h b e f o r e drinking 1 litre o f cold squash-flavoured 10% m a n n i t o l solution in f o u r divided doses b e t w e e n 18.00 and 19.00 hours on the evening before the enema. After this the patient should c o n t i n u e to drink clear fluids, b u t take no f o o d , until the e x a m i n a t i o n is over. The e x a m i n a t i o n should preferably be carried o u t in the a f t e r n o o n to allow t i m e for extra fluid resorption f r o m the b o w e l and additionally there is a t r e n d towards b e t t e r clearance w i t h the extra wait. Mannitol is reasonably palatable, and side effects such as nausea, v o m i t i n g and colic can be comparable w i t h o t h e r preparations. Its use should be avoided in those w i t h renal failure or congestive cardiac failure, w i t h the aged (over 70 years), and care is needed w i t h diabetics. The preparation can give results not significantly different f r o m a g o o d colonic lavage preparation, although it c a n n o t c o m p l e t e l y supplant a w a s h o u t facility which is still n e e d e d for selected cases.

DISCUSSION Vomiting was the most undesirable side effect o f rnannitol and could n o t be reduced in the trials b e l o w an incidence o f 10%. A l t h o u g h v o m i t i n g was usually not severe or prolonged, it had an adverse effect on the standard o f the preparation, b y reducing the antegrade washout effect if non-fluid diets were used. Metoclopramide was disappointing as an anti-emetic, but it may have had a slightly beneficial effect in reducing colic. Paradoxically the regime which produced the highest p r o p o r t i o n o f totally clear bowels was associated w i t h the highest incidence o f vomiting (25%), but this was w i t h a fluid diet regime, which has a marked effect o n results. Since this trial we have c o n t i n u e d to use manrdtol w i t h a fluid diet regime on the day before the enema, but have advised abstinence from fluids for 2 h before starting to take the mannitol. This has decreased the incidence o f vomiting to an acceptably low level. No m e t o c l o p r a rnide is given and g o o d results are being achieved. It is thought that the previous high incidence o f vomiting was due to fluid overload o f the u p p e r small bowel.

Acknowledgements. To Dr P. Ward, Consultant Radiologist, for assessing the majority of the films and encouragement, also to Drs J. F. Ratcliffe and T. E1-Husseini for their advice and interest. The trials could not have been carried out without the full cooperation of the clerical staff of the Outpatient X-ray Department who were very forebearing. Thanks are also due to Dr'D. C. Cumberland and Sister A. Herrington of the X-ray Department, Northern General Hospital, Sheffield, for their help with the lavage trial. Dr C. D. Holdsworth, Consultant Physician at the Royal Hallamshire Hospital, is thanked for helpful advice, and Dr N. Fieller of the Department of Probability and Statistics, University of Sheffield, helped with the analysis of results.

REFERENCES Minervini, S., Alexander-Williams, J., Donovan, I. A.,Bentley, S. & Keighley, M. R. B. (1980). Comparison of three methods of whole bowel irrigation. American Journal o f Surgery, 140, 400-402. Nasrullah, S. M. & Iber, F. L. (1969). Mannitol absorption and metabolism in man. American Journal o f Medical Science, 258, 80-88. Palmer, K. R. & Khan, A. N. (1979). Oral mannitol: a simple and effective bowel preparation for barium enema. British Medical Journal, 2, 1038.