Accuracy of Artificial Intelligence in Estimating Best-Corrected Visual Acuity From Fundus Photographs in Eyes With Diabetic Macular Edema

Importance

Best-corrected visual acuity (BCVA) is a measure used to manage diabetic macular edema (DME), sometimes suggesting development of DME or consideration of initiating, repeating, withholding, or resuming treatment with anti-vascular endothelial growth factor. Using artificial intelligence (AI) to estimate BCVA from fundus images could help clinicians manage DME by reducing the personnel needed for refraction, the time presently required for assessing BCVA, or even the number of office visits if imaged remotely.

Objective

To evaluate the potential application of AI techniques for estimating BCVA from fundus photographs with and without ancillary information.

Design, Setting, and Participants

Deidentified color fundus images taken after dilation were used post hoc to train AI systems to perform regression from image to BCVA and to evaluate resultant estimation errors. Participants were patients enrolled in the VISTA randomized clinical trial through 148 weeks wherein the study eye was treated with aflibercept or laser. The data from study participants included macular images, clinical information, and BCVA scores by trained examiners following protocol refraction and VA measurement on Early Treatment Diabetic Retinopathy Study (ETDRS) charts.

Main Outcomes

Primary outcome was regression evaluated by mean absolute error (MAE); the secondary outcome included percentage of predictions within 10 letters, computed over the entire cohort as well as over subsets categorized by baseline BCVA, determined from baseline through the 148-week visit.

Results

Analysis included 7185 macular color fundus images of the study and fellow eyes from 459 participants. Overall, the mean (SD) age was 62.2 (9.8) years, and 250 (54.5%) were male. The baseline BCVA score for the study eyes ranged from 73 to 24 letters (approximate Snellen equivalent 20/40 to 20/320). Using ResNet50 architecture, the MAE for the testing set (n = 641 images) was 9.66 (95% CI, 9.05-10.28); 33% of the values (95% CI, 30%-37%) were within 0 to 5 letters and 28% (95% CI, 25%-32%) within 6 to 10 letters. For BCVA of 100 letters or less but more than 80 letters (20/10 to 20/25, n = 161) and 80 letters or less but more than 55 letters (20/32 to 20/80, n = 309), the MAE was 8.84 letters (95% CI, 7.88-9.81) and 7.91 letters (95% CI, 7.28-8.53), respectively.

Conclusions and Relevance

This investigation suggests AI can estimate BCVA directly from fundus photographs in patients with DME, without refraction or subjective visual acuity measurements, often within 1 to 2 lines on an ETDRS chart, supporting this AI concept if additional improvements in estimates can be achieved.

Correlation of Change in Macular Thickness With Change in Visual Acuity in Diabetic Macular Edema: Post Hoc Analysis of VISTA and VIVID Trials

Purpose: To assess the correlation between the change in central subfield thickness (CST) and change in best-corrected visual acuity (BCVA) in eyes with diabetic macular edema (DME) treated with fixed-dosing intravitreal aflibercept injection (IAI). Methods: This post hoc analysis of the VISTA and VIVID randomized controlled clinical trials, in which 862 eyes with central-involved DME were randomly assigned to IAI 2 mg every 4 weeks (2q4; 290 eyes), IAI 2 mg every 8 weeks after 5 initial monthly doses (2q8; 286 eyes), or macular laser (286 eyes) and followed through 100 weeks. Correlations between the change in CST and change in BCVA from baseline to weeks 12, 52, and 100 were assessed using the Pearson correlation. Results: The respective correlations ( r [95% CI]) at weeks 12, 52, and 100 were −0.39 (−0.49 to −0.29), −0.27 (−0.38 to −0.15), and −0.30 (−0.41 to −0.17) in the 2q4 arm and −0.28 (−0.39 to −0.17), −0.29 (−0.41 to −0.17), and −0.33 (−0.44 to −0.20) in the 2q8 arm. Linear regression analysis of the correlation at week 100, adjusted for relevant baseline factors, showed CST changes accounted for 17% of the variance in BCVA changes; every 100-µm decrease in CST was associated with a 1.2-letter increase in BCVA ( P = .001). Conclusions: Correlations between the change in CST and change in BCVA after 2q4 or 2q8 fixed-dosing IAI for DME were modest. Although a change in CST might be important in determining the need for antivascular endothelial growth factor for DME at follow-up, it was not a good surrogate for VA outcomes.

Correlation between change in central subfield thickness and change in visual acuity in macular edema due to retinal vein occlusion: post hoc analysis of COPERNICUS, GALILEO, and VIBRANT

Purpose

Assess correlation between change in central subfield thickness (CST) and change in best-corrected visual acuity (BCVA) in eyes with macular edema due to retinal vein occlusion (RVO) that received intravitreal aflibercept injections (IAI).

Methods

Post hoc analysis of COPERNICUS and GALILEO trials for CRVO and VIBRANT trial for BRVO with relationships determined using Pearson correlation coefficient.

Results

In COPERNICUS, correlations (r) between change in CST and change in BCVA from baseline at weeks 12, 24, 52, and 100 were −0.36 (95% CI: −0.52, −0.18; P < 0.001), −0.38 (95% CI: −0.53, −0.20; P < 0.001), −0.44 (95% CI: −0.58, −0.27; P < 0.001), and −0.41 (95% CI: −0.56, −0.23; P < 0.001), respectively. CST changes accounted for only 21% of the variance in BCVA changes; every 100-µm decrease in CST was associated with a 2.1-letter increase in BCVA (P = 0.003). Similar findings were noted for GALILEO (r, −0.45 to −0.23) and VIBRANT (r, −0.36 to −0.32) trials.

Conclusion

In eyes treated with IAI for macular edema due to RVO, correlation between change in CST and change in BCVA was weak to moderate. While change in CST may be helpful in determining the need for anti-VEGF therapy, these findings do not support using changes in CST as a surrogate for changes in visual acuity outcomes.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00417-022-05697-4.

Correction to: Prediction of unfavorable outcomes in cryptococcal meningitis: results of the multicenter infectious Diseases International Research Initiative (ID-IRI) cryptococcal meningitis study

In the original version of this article, Mustafa Sunbul was not included in the list of authors for this article. The name has been added accordingly.

Coadministration of Lansoprazole and Naproxen Does Not Affect the Pharmacokinetic Profile of Methotrexate in Adult Patients With Rheumatoid Arthritis

Drugs prescribed for rheumatoid arthritis are often associated with gastrointestinal toxicity, and proton pump inhibitors may be coadministered for gastroprotection. In this open-label study, the effect of lansoprazole 30 mg qd and naproxen 500 mg bid on the pharmacokinetic profile of methotrexate was investigated. Twenty-seven adult rheumatoid arthritis patients on stable oral methotrexate doses (7.5-15 mg/week) for a minimum of 3 months were enrolled. Methotrexate pharmacokinetics were assessed on days -1 (methotrexate alone) and 7 (methotrexate with lansoprazole and naproxen). Pharmacokinetics of methotrexate and 7-hydroxymethotrexate were not altered by coadministration of methotrexate with lansoprazole and naproxen; point estimates and 90% confidence intervals for the peak plasma concentration and area under the plasma concentration-time curve of methotrexate and 7-hydroxymethotrexate were within the 0.80 to 1.25 boundaries. Therefore, coadministration of naproxen and lansoprazole for 7 days does not affect the pharmacokinetic profile of low doses of methotrexate.

Pharmacokinetics and pharmacodynamics of lansoprazole in children 13 to 24 months old with gastroesophageal reflux disease

OBJECTIVES

To evaluate the pharmacokinetics and pharmacodynamics of lansoprazole in children between 13 and 24 months of age with gastroesophageal reflux disease (GERD).

METHODS

From the population of 66 children with symptomatic GERD, erosive esophagitis (> or = grade 2) or esophageal pH < 4 for > 4.2% of the 24-h period who participated in a phase I/II, open-label, multicenter (11 sites) US study, a subanalysis of 8 toddlers between 13 and 24 months of age was performed. All children were treated, based on body weight, with lansoprazole 15 mg once daily for 8 to 12 weeks. If a child were still symptomatic after 2 weeks of treatment, then the dose of lansoprazole could be increased to twice daily at the discretion of the investigator. Pharmacokinetic parameters were assessed at day 5. Twenty-four-hour median intragastric pH and the percentage of time intragastric pH > 3 or > 4 were assessed at baseline and at day 5 of treatment. Symptom response was assessed by investigator interview and daily diary. Safety was monitored by physical examinations including vital signs, adverse event assessments and laboratory evaluations.

RESULTS

Pharmacokinetic analysis of 5 children found a mean time to reach maximum concentration of 1.4 h, maximal plasma concentrations of 894 ng/mL, area under the concentration time curve of 1906 ng * h/mL and a half-life of 0.66 h. Significant (P < or = 0.027) increases from baseline to day 5 were observed in mean 24-h intragastric pH (2.76-3.52) and the percentages of time pH were > 3 (29.46%-55.36%) and pH was > 4 (16.96%-40.77%). Six of the 8 children had improvement in their overall GERD symptom severity on the basis of investigator assessment, and a reduction was seen in the percentage of days with moderate, severe or very severe GERD symptoms compared with baseline. The dosage of lansoprazole was increased in 3 of the 8 children. Median fasting serum gastrin level increased from 65.0 pg/mL at baseline to 136.5 pg/mL at the final visit. Treatment-related events were mild constipation (1 subject) and mild diarrhea (1 subject).

CONCLUSIONS

Although larger studies are needed to confirm these results, lansoprazole displays pharmacokinetic and pharmacodynamic parameters in children between 13 and 24 months of age that are similar to those results observed in older children as well as adults.

Presenting symptoms of nonerosive and erosive esophagitis in pediatric patients

Children and adolescents with symptomatic gastroesophageal reflux disease (GERD) and erosive esophagitis (EE) of grade >/=2 (n=45) or nonerosive esophagitis (NEE) (n=45) were assessed to determine the relationship between presenting symptoms, esophagitis severity, and patient age. Overall, regurgitation/vomiting, abdominal pain, and cough were the most frequent symptoms. The prevalence and severity of anorexia/feed refusal was significantly greater in EE versus NEE children; this symptom was also significantly more prevalent in younger (1-5 years) children (both NEE and EE groups) compared to older children. Cough was significantly less severe in NEE adolescents than in younger children. Cough, anorexia/feed refusal, and regurgitation/vomiting were more severe and heartburn was less severe in EE children aged 1-5 years compared with older patients. In conclusion, GERD in children manifests differently than that in adults and symptoms vary with patient age. Symptoms were not predictive of presence or lack of mucosal damage.

Dose-response evaluation of the antisecretory effect of continuous infusion intravenous lansoprazole regimens over 48 h

BACKGROUND

Attainment of intragastric pH < 6.0 may require high-dose continuously infused proton pump therapy.

AIM

To assess the pharmacokinetic and pharmacodynamic dose-responses of continuous infusion regimens of lansoprazole.

METHODS

Healthy adult subjects were assigned to lansoprazole 60-mg intravenous bolus, followed by 6-mg/h continuous infusion; a 90-mg intravenous bolus followed by 6-, 7.5-, or 9-mg/h continuous infusion; or placebo.

RESULTS

Mean intragastric pH values for lansoprazole regimens ranged from 4.8 to 5.2 (0-24 h), 5.5 to 6.0 (>24 to 48 h) and 5.2 to 5.6 (0-48 h). Within these three intervals, the percentages of time intragastric pH exceeded 4, 5 and 6 ranged from 65% to 96%, 54% to 88% and 30% to 61% respectively. Pharmacokinetic parameters were dose-independent with steady-state plasma concentrations achieved within 6-12 h postdose and maintained over 48 h. The mean systemic clearance of lansoprazole was lower in CYP2C19 heterozygous metabolizers than in homozygous extensive metabolizers (9.2 vs. 16.5 L/h), with substantial variability resulting in overlapping ranges of clearance values for both subpopulations. All lansoprazole regimens were well-tolerated.

CONCLUSIONS

Lansoprazole administered as a 60-mg intravenous bolus followed by 6-mg/h continuous infusion produced intragastric pH effects comparable with those of higher dosage regimens.

Lansoprazole regimens that sustain intragastric pH > 6.0: an evaluation of intermittent oral and continuous intravenous infusion dosages

BACKGROUND

Orally and intravenously administered proton pump inhibitors have been shown to reduce rebleeding rates, surgery and transfusion requirement.

AIM

To compare lansoprazole intravenous and orally disintegrating tablet (Prevacid SoluTab) regimens with a pantoprazole intravenously administered regimen in sustaining intragastric pH >6.0.

METHODS

Two similarly designed three-way, randomized crossover studies each enrolled 36 Helicobacter pylori-negative healthy volunteers. Study 1 regimens included intravenously administered bolus followed by 24-h continuous infusion (lansoprazole 90 mg, 6 mg/h; lansoprazole 120 mg, 6 mg/h; pantoprazole 80 mg, 8 mg/h). Study 2 regimens included intravenous bolus followed by lansoprazole orally disintegrating tablet or intravenous continuous infusion for 24 h (lansoprazole 90 mg, lansoprazole orally disintegrating tablet 60 mg every 6 h; lansoprazole 120 mg, 9 mg/h; pantoprazole 80 mg, 8 mg/h). Percentage of time pH >6.0 was assessed with 24-h intragastric pH monitoring.

RESULTS

All regimens produced comparable gastric acid suppression. In both studies, regimens superior to pantoprazole included lansoprazole 90 mg, 6-mg/h; lansoprazole 90 mg, lansoprazole orally disintegrating tablet 60 mg q.d.s. and lansoprazole 120 mg, 9 mg/h (P < or = 0.013). The lansoprazole 120-mg, 6-mg/h regimen (P = 0.082) was not superior to pantoprazole in percentage of time intragastric pH >6.0. Mild reaction at the intravenous injection site was the most frequently reported adverse event.

CONCLUSIONS

The intravenous bolus and continuously infused lansoprazole or intravenous bolus and intermittent lansoprazole orally disintegrating tablet regimens are as effective as intravenous pantoprazole in sustaining intragastric pH >6.0.

Ulcer recurrence in high-risk patients receiving nonsteroidalanti-inflammatory drugs plus low-dose aspirin: results of a post HOC subanalysis

BACKGROUND

Concomitant aspirin use is a risk factor for nonsteroidal anti-inflammatory drug (NSAID)-associated upper gastrointestinal toxicity. In high-risk individuals, such as those with a history of NSAID-related gastric ulcer bleeding, gastroprotective therapy with a proton pump inhibitor has been reported to reduce the risk of recurrent aspirin-associated gastroduodenal ulcer bleeding.

OBJECTIVE

This analysis compared the efficacy of misoprostol, lansoprazole, and placebo in reducing the risk of gastric or duodenal ulcer recurrence in patients taking NSAIDs and low-dose aspirin.

METHODS

This post hoc subanalysis was based on a previous multicenter, prospective, randomized, double-blind, placebo-controlled, 12-week study in patients who had a history of gastric ulcer, were Helicobacter pylori negative, required chronic NSAID therapy, and were free of gastric or duodenal ulcer on baseline endoscopy. The study treatments were misoprostol 200 microg QID or lansoprazole 15 or 30 mg OD. The subanalysis included data from patients in the intent-to-treat cohort who took aspirin at an amount <or=325 mg/d. The end point was the cumulative rate of gastric ulcers, as assessed by serial endoscopy at 4, 8, and 12 weeks.

RESULTS

Of 535 intent-to-treat patients from the primary study, 70 (40 men, 30 women; mean [SD] age, 64.7 [10.0] years; age range, 40-83 years) met the criteria for inclusion in the subanalysis. The proportions of patients who were free of gastric ulcers at the end of 12 weeks were 96% in the misoprostol group, 93% in the lansoprazole 15-mg group, 100% in the lansoprazole 30-mg group, and 35% in the placebo group (P <or= 0.008, each active treatment vs placebo). Adverse events considered possibly or probably related to treatment occurred in 5 (20.0%) misoprostol recipients (4 episodes of diarrhea, 1 episode of abdominal pain), 1 (14.3%) recipient of lansoprazole 30 mg (1 episode of pharyngitis), and 3 (13.6%) placebo recipients (1 episode each of abdominal pain, palpitations, and dyspepsia).

CONCLUSIONS

In this subgroup analysis in patients at high risk for recurrence of gastric ulcer, use of cotherapy with misoprostol 200 microg QID or lansoprazole 15 or 30 mg OD significantly lowered the risk for gastric ulcer recurrence.

Taste Comparisons for Lansoprazole Strawberry-Flavoured Delayed-Release Orally Disintegrating Tablet and Ranitidine Peppermint-Flavoured Syrup in Children

OBJECTIVE

To compare the flavour and taste preferences of two acid-inhibitory therapies in children.

SUBJECTS AND SETTING

104 (52 male and 52 female) healthy children aged 6-11 years participated in this phase IV single-centre, taste-test study after parental consent was obtained.

METHODS

Children were divided into two groups based upon age: group 1 included children aged 6-8 years and group 2 included children aged 9-11 years. Within each group an equal number of male and female subjects were recruited. Within each of the four strata defined by group and sex, an equal number of subjects were randomised to taste the strawberry-flavoured lansoprazole 15mg orally disintegrating tablet or the ranitidine 75mg/5mL peppermint-flavoured syrup samples in position 1. In group 1, the lansoprazole 15mg delayed-release orally disintegrating tablet was dispersed in 5mL of water, while in group 2, children gently rolled the tablet on the tongue until dissolution, before swallowing the particles. Children given the dose of lansoprazole dispersed in water (group 1) and the ranitidine dose (groups 1 and 2) were to taste it, swish it in their mouth for up to 10 seconds, and then swallow it. Children were given ambient temperature water and unsalted crackers to cleanse the palate during a 10-minute break between tastings.

MAIN OUTCOME MEASURES

After each tasting, children rated their degree of liking on a five-point facial hedonic scale (5 = like very much, 1 = dislike very much). Product preference was recorded after the tasting of both samples.

RESULTS

Among group 1 and group 2 participants, 86.5% (45/52) and 90.4% (47/52) of children, respectively, 'liked' ('like a little' or 'like very much') the strawberry-flavoured lansoprazole orally disintegrating tablet. The proportion of children who 'liked' the peppermint-flavoured ranitidine syrup was lower than the proportion who liked lansoprazole, and the proportions were similar between the groups: 13.5% (7/52) in group 1 and 9.6% (5/52) in group 2. Children in both groups preferred the strawberry-flavoured lansoprazole delayed-release orally disintegrating tablet: 92% (95% CI 81.1, 97.8; p < 0.001) of those in group 1 and 98% (95% CI 89.7, 100.0; p < 0.001) of those in group 2.

CONCLUSION

After tasting both products, >92% of children aged from 6-11 years preferred the strawberry-flavoured lansoprazole delayed-release orally disintegrating tablet, either dissolved in a small amount of water or allowed to dissolve on the tongue, over the peppermint-flavoured ranitidine syrup.

Comparison of the pharmacokinetics of lansoprazole 15- and 30-mg sachets for suspension versus intact capsules

OBJECTIVE

The pharmacokinetic profiles of single doses of lansoprazole 15- and 30-mg sachets for suspension were compared with those of corresponding doses of lansoprazole oral capsules.

METHODS

Healthy adult male and female subjects were randomized (1:1 ratio) into 2 Phase 1, open-label, single-dose, 2-sequence, 2-period complete crossover studies. In the first study, each subject received 1 lansoprazole 15-mg sachet mixed with water and 1 lansoprazole 15-mg oral capsule; in the second study, each subject received 1 lansoprazole 30-mg sachet mixed with water and 1 lansoprazole 30-mg oral capsule. Administration of the 2 formulations was separated by a washout period of > or =7 days. Blood samples were collected before and after each administration to assess the pharmacokinetic parameters of lansoprazole and bioequivalence between suspension and capsule.

RESULTS

Thirty-six subjects (19 males, 17 females) with a mean (SD) age of 32.0 (9.6) years and mean (SD) body weight of 68.6 (10.5) kg received lansoprazole 15 mg. Thirty-six subjects (22 males, 14 females) with a mean (SD) age of 38.0 (8.3) years and mean (SD) body weight of 75.1 (9.7) kg received lansoprazole 30 mg. The pharmacokinetic parameters of the 15- and 30-mg lansoprazole sachets for suspension were similar to those of the corresponding doses of the oral capsules. The mean (SD) values for C(max) and AUC from time 0 to infinity (AUC(0-infinity) for the lansoprazole 15-mg sachet (591.9 [242.3] ng/mL and 1614 [2065] ng.h/mL, respectively) did not differ significantly from those for the lansoprazole 15-mg capsules (578.6 [275.2] ng/mL and 1620 [2290] ng.h/mL, respectively). These parameters also did not differ significantly between the lansoprazole 30-mg sachet and 30-mg capsule: mean (SD) C(max), 1103 (428.3) and 1077 (465.6) ng/mL, respectively; mean (SD) AUC(0-infinity), 2655 (1338) and 2669 (1311) ng.h/mL, respectively. The 90% Cls for C(max) and AUC(0-infinity) ratios were contained within the 0.80 to 1.25 equivalence range, supporting bioequivalence.

CONCLUSIONS

These findings suggest that the 15- and 30-mg lansoprazole sachets for suspension are bioequivalent to the corresponding doses of oral capsules. The sachet for suspension may provide an alternative route of administration to patients who have difficulty swallowing solid oral formulations.

Pharyngeal repair after laryngopharyngectomy: 4-year experience

The most common forms of surgical intervention carried out at the National Cancer Institute of Cairo for patients initially seen with hypopharyngeal carcinoma are (1) laryngopharyngectomy and pharyngeal repair using the pectoralis major myocutaneous flap (PMMF) and (2) extrapleural total pharyngolaryngoesophagectomy with gastric transposition (GT) and pharyngogastrostomy. The advantages and disadvantages with both techniques will be discussed on the basis of results of 69 patients with different tumor distributions over a 4-year period.

Postoperative radiotherapy of carcinoma in bilharzial bladder using a three-fractions per day regimen

Patients with T3 bladder cancer who survived surgery and proved to have P3a, P3b or P4a tumors were randomized to either no further treatment (61 patients) or postoperative total pelvic irradiation (55 patients). A three-fraction per day regime was adopted with a dose per fraction of 125 cGy and an interval of 3 h between fractions. The total dose amounted to 3750 cGy divided into 30 fractions over 12 days. Patients of the postoperative radiotherapy group were re-randomized to radiotherapy alone or radiotherapy plus misonidazole (MISO) in a daily dose of 1 g/m2 given orally 2 h before the first daily fraction. The 2-year disease-free survival rate in the cystectomy alone group was 33 +/- 6% compared to 65 +/- 6% in the postoperative radiotherapy group. The therapeutic benefit applied to the two cell types, all histological grades and stages and to patients with or without nodal metastases. The benefit of postoperative irradiation was also verified by the Cox's multivariant analysis which adjusts for the relative representation of the important prognostic factors particularly pathological stage and nodal involvement. MISO did not seem to add to the therapeutic gain. No late complications were encountered in the wall of the rectum, small bowel or uretero-intestinal anastomotic sites. This is suggested to be due to the small dose per fraction used. However, early small bowel reactions were dose-limiting.

Primary closure of the perineum and vacuum drainage after abdominoperineal excision

From retrospective study of the case records of 62 consecutive patients from whom 58 underwent abdominoperineal rectal excision for malignant or inflammatory bowel disease, it was found that management with primary closed perineum and closed pelvic peritoneum with closed vacuum drainage results in 84% primary perineal healing and short hospitalization. Perineal wounds were healed in 93% of patients within one month. Adequate exclusion of three sepsis promoting factors, e.g. dead space, devitalized tissue and micro-organisms should preclude perineal sepsis. The need for precise positioning of a large bore vacuum drain, Ch. 27, through a preoperatively determined stab-wound on the medial side of the thigh is stressed. To guarantee optimal drainage a three-stage procedure is required in a one-team management. The procedure is well tolerated as a palliation; patients with pre-existing massive perineal sepsis should be excluded.

The risk of central venous catheter-related sepsis in patients with surgical infections

A prospective study on the incidence of central venous catheter-related sepsis (CRS) in patients with or without concomitant bacterial infections was performed. In 194 patients, 315 central venous catheters were placed, mainly for parenteral nutrition and CVP measurement. Catheter sepsis in patients with surgical sepsis was 4.0% compared with 3.6% in the non-infected group. CRS occurred exclusively in patients with an open distant infection. Subdivision of the group of patients with infection showed that those with 'open' infection had significantly more CRS than those without open wounds. The results strongly suggest an almost exclusive exogenous route for catheter contamination. It is concluded that risk of catheter sepsis is not enhanced by the presence of concomitant infections located elsewhere. The low incidence and relative benign course of catheter-related sepsis in seriously ill patients with a surgical infection tends to widen the indication for central venous catheterisation rather than to restrict it.

Pre-operative whole-gut irrigation with mannitol

Whole-gut irrigation preceded by oral mannitol administration as a method of preoperative bowel preparation was studied in 58 patients. Our results did not confirm the advantages described in the literature. Irrigation time, irrigation fluid volume, patient acceptability and the quality of bowel preparation were not improved, compared to our earlier studies with saline alone. Fluid retention was considerably diminished after mannitol preparation. A significant increase in postoperative septic complications was observed. Disadvantages and possible hazards of mannitol administration are discussed. Mannitol cannot be recommended as a safe and useful adjunct in whole-gut irrigation.