Can post-operative antibiotic prophylaxis following elective laparoscopic cholecystectomy be completely done away with in the Indian setting? A prospective randomised study.

PREMISE AND
OBJECTIVE: : Elective laparoscopic cholecystectomy (LC) has low risk for post-operative infectious complications; still most clinicians use persistent post-operative prophylactic antibiotics out of habit, tradition, or simply as defensive practice due to evolving medicolegal implications of a large number of surgeries being showcased as daycare or next day discharge procedures. This randomised prospective trial was done to test the need for such prophylaxis in cases of elective LC in a rural/semi-urban setting.
MATERIALS AND METHODS: : Two hundred and ten successive patients undergoing elective LC were randomised into groups receiving single dose of injection ceftriaxone at the time of induction of anaesthesia, (Group A = 112 cases) and those who in addition to above received injection ceftriaxone twice daily for 2 days postoperatively (Group B = 98 cases). Post-operative infectious complications between two groups were compared for variables such as age, sex, body mass index and bile/stone spillage.
RESULTS: : There was no significant difference in surgical site infection rates between the groups for variables such as age, sex, body mass index, duration of symptoms, American Society of Anesthesiologists grade, duration of surgery and hospital stay. Intraoperative spillage of stones (9.8% [A]: 5.1% [B]) did not increase infectious complications even in the presence of positive bile culture (Group A, N = 7 vs. Group B, N = 3). An operative time of greater than 60 min was found to be associated with increased surgical site infection (P = 0. 0006).
CONCLUSION: : Single dose of ceftriaxone at the time of induction is adequate prophylaxis following elective LC even in the rural/semi-urban Indian setting and routine continued administration of antibiotic should be abandoned as it contributes to adverse reactions, drug resistance and unnecessary financial burden.

INTRODUCTION

Laparoscopic cholecystectomy (LC) is the gold standard for treating symptomatic cholecystitis and is associated with fewer complications than its open counterpart.[12] Surgical site infections (SSIs) postoperatively can lead to significant morbidity and mortality and are prevented in clean and clean-contaminated wounds by the use of antibiotic prophylaxis.[345] However, routine use of antibiotics after clean LC has been equally questioned[678910] and supported.[1112131415] Although Center for Disease Control and Prevention recommends the administration of prophylactic antibiotics in clean-contaminated surgery such as LC to reduce surgical site infections (SSI),[3] more recent meta-analyses have concluded that antibiotic prophylaxis is not warranted in low-risk patients undergoing LC. Current consensus seems to be to not use antibiotics in patients undergoing LC for low- or moderate-risk groups, owing to the modest risk of developing an SSI and costs to the health-care system.[16] Simultaneously, there is increasing awareness and efforts to reduce inappropriate antibiotic use with consequent multidrug microbial resistance and problems such as increasing rates of Clostridium difficile infection.[1718] In spite of the above recommendations, between 20% and 80% of patients undergoing LC with a low and moderate risk of SSI still receive antibiotics in various studies.[1617181920]

With this knowledge, we conducted this study to find that whether these recommendations are applicable and feasible in the resource sensitive Indian setup where finances and follow-up needs may affect treatment outcomes.

Aim and research question

The aim and research question is single dose antibiotic prophylaxis at the time of induction of anaesthesia safe and effective in elective LC in a semi-urban setting?

Can persistent postoperative antibiotic prophylaxis after elective LC be done away with completely in a semi-urban setting?

MATERIALS AND METHODS

This prospective randomised study was conducted at School of Medical Sciences and Research, Sharda Hospital, Greater Noida. A total of 210 consecutive patients undergoing LC from July 2016 to January 2017 after valid informed consent and due clearance from the Ethical Committee were included as governed by inclusion/exclusion criteria as follows:

Inclusion criteria

All cases admitted for routine elective LC. Spillage of bile and inadvertent perforation of gallbladder included.

Exclusion criteria

Cholangitis, obstructive jaundice

Evidence of biliary pancreatitis

History of oral/parenteral antibiotic intake in past 2 weeks

History of endoscopic retrograde cholangiopancreatography/endoscopy in the past month

Patients with cardiac prosthesis

Empyema of gall bladder

Diabetes mellitus.

Patients were randomised to two groups: Group A was administered injection ceftriaxone (1 g) intravenously at the time of induction and Group B received in addition injection ceftriaxone (1 g) intravenously postoperatively twice daily for 2 days. Simple randomisation into Group A and B was done by the anaesthetist picking a chit to assign the case to either group, without the knowledge of the treating surgical team.

After preparing the part with 10% povidone-iodine, LC was completed using three- or four-port technique. After pneumoperitoneum creation with Veress needle, an infraumbilical 10 mm port was made and subsequently 2/3 ports as need to be under vision of 10 and 5 mm size at epigastrium, above fundus of gallbladder, in the right paraumbilical region in midclavicular line, respectively. Perforation of gallbladder and spillage of bile was managed by removal of spilled stones and peritoneal toilet with 1 l of normal saline. Gallbladder was removed from the epigastric port in an endobag. In case of spillage, the toileting fluid was sucked and sent for culture. Ports were closed by monofilament 3-0 non-absorbable sutures. Record was made of duration of surgery, post-operative fever, tachycardia, port site discharge or inflammation, intra-abdominal collection and hospital stay duration. Patients were discharged on the second post-operative day unless there were any signs such as abdominal pain, tenderness, distension, bile leak, deranged vitals or patient unwillingness to do so. Following discharge, the patients were assessed for any evidence of SSI or deep intra-abdominal infection at 15 days and at 1 month.

Assuming the margin of error that could be accepted as 5%, with a confidence level of 90% and population size of 325 (the number of cases who undergo LC typically in a year by our unit). Assuming the response distribution to be 50%, with the above assumptions, the sample size calculated is 148, using the following formula:

Sample size n and margin of error E are given by

x = Z (c/100)2r (100 − r)

n = N x/([N − 1]E2 + x)

E = √(N – n) x/n (N – 1)

Where N is the population size, r is the fraction of responses that you are interested in and Z (c/100) is the critical value for the confidence level c. This calculation is based on the normal distribution and assumes that there are more than thirty samples and a power of 80.

Statistical methods

Chi-square test, Fisher's exact test and Pearson's Chi-square test were done for statistical analyses as appropriate using IBM SPSS Statistics for Windows, version 23 (IBM Corp., Armonk, N.Y., USA). P < 0.05 was considered statistically significant.

RESULTS

Analyses of the data showed that there was no significant variation between the two groups with regard to variables such as age, sex, body mass index (BMI), duration of symptoms, American Society of Anesthesiologists score, duration of surgery, hospital stay and SSI [Table 1]. There was no case with deep intra-abdominal infection requiring intervention. Bile/stone spillage during surgery did not affect outcome significantly between the two groups (P = 0.19). Two patients (one from each group) developed significant bile leak (>200 ml/day lasting more than 7 days), requiring switching of one Group A case to Group B. Repeat ultrasound and post-operative magnetic resonance imaging in both cases showed normal size common bile duct without any injury. Both cases responded to conservative management with spontaneous resolution of leak in 9 and 13 days, respectively.

Table 1

Patient characteristics and variables

Escherichia coli and Staphylococcus sp. were most commonly cultured types in spilled bile (16 cases), only 10 of these 16 cases had positive cultures, persistence of appropriate group protocol did not affect outcome [Tables 1 and 2]. Two male patients from Group B developed dysuria on 3rd day with urine examination showing 8–10 pus cells/hpf with sterile culture (not significant, P = 0.21), managed with the continuation of the group protocol.

Table 2

Bacteria isolated from spilled gall bladder bile

Overall, there was no significant difference in between both groups except when comparing stratified variables as risk factors for infectious outcomes [Table 3], where it was demonstrated that an operative time of more than 60 min was associated with significant SSI (P = 0.0006). One patient from Group B developed drug reaction to ceftriaxone (rashes and papules with dry mouth on the second post-operative day), hence drug was stopped. Two patients from Group A and one from Group B were lost to follow-up after the 15th day visit.

Table 3

Comparison of the two groups with stratification of risk variables for infectious complications at 1 month follow-up after laparoscopic cholecystectomy

DISCUSSION

In our study, overall rate of infection was 2.3% that compares favourably to rates of 0.4%–6.3% from similar studies.[678101316] Although the infection rate was higher in a persistent antibiotic group versus single dose group (3% vs. 1.8%), it was not statistically significant, which is similar to the above quoted studies. None of the patients in either group developed any deep intra-abdominal or systemic infections. Earlier studies by Lippert and Gastinger, Uchiyama et al.,[1112131415] advocated that both laparoscopic and conventional open cholecystectomy are performed with adequate perioperative antimicrobial prophylaxis because patients receiving prophylaxis fared significantly better than those with no prophylaxis in terms of the rate of post-operative infections, other complications, reoperation and mortality. However, others have opined the opposite that prophylactic antibiotics neither affect the development nor the course of infectious complications following LC as the infection rate is very low.[678910] Our study demonstrated the same with no significant complications in Group A.

Spillage of stones/bile leak occurs during LC due to traction, dissection and manipulation during extraction of gall bladder and is seen in 11%–30% cases.[691215] The bacterial culture positivity rates in gallbladder stones have been shown to be between 10% and 42%.[679121314] Shindholimath et al.[13] concurred that bactibilia was the most important predictor of wound infection in low-risk patients undergoing elective LC and recommended prophylactic antibiotic coverage for such cases. Dervisoglou et al.[14] and Darzi et al.[21] concluded that both positive bile culture and intraoperative gallbladder rupture were strongly associated with the development of SSI. They also found in their study that SSI was caused by exactly the same pathogens found in intraoperative cultures. However, in our study, we found that positive bile culture and perforation of gallbladder and bile leak/spillage of stones did not lead to increased infectious complications [Table 1]. None of the 16 (7.6%) cases with spillage of bile/stones and those with positive irrigant fluid culture (10) developed SSI [Table 3]. Other newer studies have also concluded the same that SSIs are not related to bile culture, rupture of the gallbladder, or spillage of gallbladder stones or bile.[46791022] We believe thorough peritoneal toilet and suction of irrigant fluid and isolation of specimen in endobag prevents the development of SSI even in spillage of positive bile or stones.

Stratification of accepted independent risk factors for complications in LC, like increased age, male sex, number of ports, BMI, did not show increased complications in our study between the two groups [Table 3]. As the mean BMI for the two groups was similar and most patients did not have grossly abnormal BMI, stratification for same was not done in analyses. Other studies have also not found differences in complications between normal-weight and overweight patients.[232425] There was a significant association of infectious complications with increased operative time [Table 3, P = 0.0006] in our study which is similar to accepted findings by other authors.[26] A recent comprehensive meta-analyses and trial by Matsui et al.[27] showed a reduction in infectious complications with three perioperative doses of prophylactic antibiotics in elective low-risk LC and that eliminating the use of prophylactic antibiotics did not result in a cost reduction due to trials being statistically underpowered, no antibiotic at all in one group, lack of 100% follow-up. Our study does have the first and third limitations as stated above and we chose injection ceftriaxone instead of injection cefazolin as described in almost all studies as its more easily available and has a more extended spectrum.[2829] Our choice of administering antibiotic at the time of induction in one group aids the host immune system in tiding over the most likely period of infectious complications rather than completely eliminating pathogens and is a balance between no antibiotics at all and prolonged antibiotics administration. Our study shows that risk of SSI in LC is low and does not seem to be reduced by the routine use of prolonged post-operative antibiotic prophylaxis in uncomplicated cases, still most clinicians continue to persist with it out of tradition, habit or medicolegal reasons (defensive practice due to projection of LC as an early/next-day discharge surgery). As most of the study data are from the developed nations, it cannot be applied as such in countries like India due to different patient profile, social, financial variations in healthcare delivery setup. Also, there is a very real risk of adverse reactions to antibiotic use that may lead to significant morbidity and financial burden. In view of the above, we suggest that a single dose of antibiotic at induction is the most practical and effective choice of prophylaxis protocol.

CONCLUSION

Persistent post-operative antibiotic prophylaxis following uncomplicated low-risk LC is unnecessary and a single dose of antibiotic at the time of induction provides an adequate balance between prevention of infectious complications and injudicious use of antibiotics.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.