Article Text

Download PDFPDF
Drain in laparoscopic cholecystectomy in acute calculous cholecystitis: a randomised controlled study
  1. Mithun V Valappil1,
  2. Sumit Gulati1,
  3. Manish Chhabra1,
  4. Ajay Mandal1,
  5. Sanjay De Bakshi1,
  6. Avik Bhattacharyya2,
  7. Supriyo Ghatak1
  1. 1 Surgical Gastroenterology, Calcutta Medical Research Institute, Kolkata, India
  2. 2 Interventional Radiology, Calcutta Medical Research Institute, Kolkata, India
  1. Correspondence to Dr Supriyo Ghatak, Surgical Gastroenterology, Calcutta Medical Research Institute, Kolkata 700027, India; drsupriyo{at}


Background There is paucity of evidence regarding the role of drain in laparoscopic cholecystectomy (LC) in acute calculous cholecystitis (ACC), and surgeons have placed the drains based on their experiences, not on evidence-based guidelines. This study aims to assess the value of drain in LC for ACC in a randomised controlled prospective study.

Patients and methods All patients with mild and moderate ACC undergoing LC were assessed. Preoperatively, patients with choledocholithiasis, Mirizzi syndrome and biliary stent were excluded. Intraoperatively or postoperatively, patients with complications, partial cholecystectomies and malignancies were excluded. Patients were randomised using computer-generated random numbers into two groups at the end of cholecystectomy before closure. Requirement of radiologically guided (ultrasonography () or CT) percutaneous aspiration/drainage of symptomatic intra-abdominal collection or reoperation; continuation of parenteral antibiotics beyond 24 hours or change in antibiotics empirically or based on peritoneal fluid culture sensitivity; requirement of postoperative USG or CT scan based on postoperative clinical course; wound infection rates; postoperative pain using numeric rating scale at 6 and 24 hours; and the duration of hospital stay in both groups were noted.

Results Forty-two out of 50 consecutive patients were randomised into two equal groups. Pain score at 6 and 24 hours was less in patients without drain. All other complication rates and duration of stay were similar in both groups.

Conclusions Drains should not be placed routinely after LC in ACC as it increases pain and does not help in detecting or decreasing complications.

  • acute cholecystitis
  • gallstones
  • laparoscopic cholecystectomy
  • drain

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Most surgeons use routine subhepatic drain for the fear of bile leak and bleeding after laparoscopic cholecystectomy (LC). But it has been conclusively proven that drainage after LC in biliary colics or uncomplicated cholelithiasis is unnecessary.1 2 However, there is paucity of evidence regarding the role of drain in acute calculous cholecystitis (ACC) after LC, and surgeons have placed the drains based on their experiences and not on evidence-based guidelines. This study aimed to assess the value of drain in LC for ACC in a randomised controlled prospective study. Requirement of radiologically guided percutaneous aspiration of intra-abdominal collection or reoperation; continuation of parenteral antibiotics beyond 24 hours or change in antibiotics; requirement of postoperative ultrasonography (USG) or CT scan; wound infection rates; postoperative pain using numeric rating scale (NRS) at 6 and 24 hours; and the duration of hospital stay in both groups were noted.

Patients and methods

The study was conducted on patients undergoing LC with mild and moderate ACC (Tokyo Guidelines 2013) in the Department of Surgical Gastroenterology, CMRI Hospital, from March 2016 to May 2018 after taking institutional ethics committee clearance. Patients with choledocholithiasis, Mirizzi syndrome, biliary stent, cholangitis and gall bladder (GB) malignancy; patients who underwent partial or open cholecystectomy; and patients in whom drains were placed in anticipated or actual bile duct injury were excluded.

At the time of starting the trial we could not find any reliable study in the literature. We did this study as a pilot project and consequently the study was not powered to calculate sample size. Our sample size was based on the number of patients with acute cholecystitis who underwent LC in our institute in the last 1 year (n=50).

Consenting patients meeting the inclusion criteria were randomised using computer-generated random numbers online (, using QuickCalcs generated on 6 September 2016) and sealed envelopes were opened at the operation theatre after the removal of the GB. Intravenous cefuroxime 1.5 g was given at the time of induction of anaesthesia. The dose was repeated twice postoperatively eight hourly. The classical four-port LC was done. Port sites were infiltrated with 0.25% injection of bupivacaine before the incision. GB was retrieved through the infraumbilical port. Bile was sent for culture and sensitivity (c/s), taken from GB before or after extraction. Thorough irrigation with 250–500 mL of normal saline was done in all patients after the GB was extracted. A 20F closed tube drain was placed in subhepatic space through the lateralmost port in group A patients. Postoperatively, all patients were given oral paracetamol 650 mg tablets, thrice daily starting 6 hours after surgery for 24 hours and then as and when required. The requirement of an additional intravenous paracetamol 1 g was recorded. The pain was assessed using NRS at 6th and 24th hours postoperatively. The amount of drainage was measured on first postoperative day and the drain was removed if the effluent was serous and <50 mL in the last 24 hours.

Postoperatively, USG of abdomen was done, if any of the following two criteria were positive at 24 hours: fever > 38.34 Celsius, continuous abdominal pain, abdominal distension, tachycardia (>90/min) and NRS>5.

Postoperatively, intravenous antibiotics were continued for 3 days, if any of the above-mentioned signs or symptoms are present at 24 hours. Antibiotics were changed either, if the patient was not improving with the current drug or, depending on the c/s result of the aspirate.

If there was a collection on USG, aspiration was done and the treatment was decided on the nature of the aspirate. If the nature was serous, patients were treated with needle aspiration alone. If the nature is thick, like blood or pus, it was drained using a pigtail catheter of size 12F. Second aspiration was tried if the collection recurs. Third recurrence was to be treated with surgical drainage. The aspirate was sent for c/s.

If patients do not show any collection on USG and the above-mentioned clinical signs were positive, a CT scan of abdomen was obtained. Drainage policy was followed same as in USG. Duration of stay was calculated from the day of surgery to the day of discharge.

Statistical analysis

The primary endpoint in this study was to evaluate if there was any difference in radiological or surgical reintervention rate. The secondary endpoints were requirement or change of antibiotics beyond 24 hours, requirement of USG or CT scan, difference in pain score (NRS) at 6 hours and 24 hours and length of stay.

Categorical variables were expressed as number and percentage of patients and compared across the two groups using Pearson’s χ2 test for independence of attributes. Continuous variables were expressed as mean±SD and compared across the two groups using unpaired t- test if the data follow the normal distribution and Mann-Whitney U test if the data do not follow the normal distribution. Statistical software SPSS V.20 was used for the analysis. An alpha level of 5% was taken as significant.


Fifty patients were eligible to participate in the study (table 1). Preoperatively, four patients were excluded due to following reasons: diagnosis of Mirizzi syndrome (n=2) and choledocholithiasis (n=2). Intraoperatively, four patients were excluded because the surgeon himself has decided to place drain due to the following reasons: gangrenous GB with bleeding from liver bed (n=1), frozen Calot’s triangle and non-identification of cystic duct (n=1), wide friable cystic duct which was sutured (n=1) and bleeding from liver bed (n=1).

Table 1

Patient characteristics and postoperative parameters

After exclusion, 42 patients were enrolled in this study. Twenty-one patients were allocated each into two groups, A and B, after randomisation using random number generator. In group A drain was placed, and in group B patients did not receive drain. All the patients were followed up for 1 week.

Fifteen out of 21 (71.43%) and 11 out of 21 (52.38%) were female in group A and group B, respectively (p=0.2). There was no difference between both groups in terms of age distribution (p=0.42) and presence of comorbid conditions (p=1).

Out of 42 patients, 38 were operated within 72 hours. Four patients were operated after 72 hours, all were diagnosed with moderate ACC. Out of these four patients, three were in drain group and one patient was in no drain group.

Thirty-four patients had mild ACC (80.95%). In these 34 patients, 19 (90.48%) were in group B and 15 (71.43%) were in group A. A total of eight patients (19.05%) were diagnosed with moderate ACC: six in group A and two in group B. There was no difference in both groups in terms of diagnosis (p=0.24).

Eight patients (19.05%) received more than three doses of antibiotics. Three patients received a total of 5 days and five patients received a total of 4 days of antibiotics (including preoperative doses). All patients were diagnosed with moderate ACC. None of the patients required change in antibiotics. There was no difference between both groups with respect to dosage of antibiotics (p=0.24)

Only one patient (2.38%) with mild ACC in group B required postoperative imaging. This patient had postoperative tachycardia and pain. A USG was done which did not reveal any collection. There was no difference in both groups in terms of postoperative imaging (p=0.31).

Ten patients (47.62%) in group A had a pain score of 5 at 6 hours whereas only three patients (14.29%) in group B had a score of 5 at 6 hours (p=0.006). Five patients in drain group and four patients in no drain group had received additional doses of analgesics. Ten patients (47.62%) in group A had a score of 4 at 24 hours, whereas only two patients (9.52%) out of 21 in group B had a score of 4 at 24 hours (p=0.003). None of the patients received further additional doses of analgesics after 6 hours.

The mean duration of stay was not different in two groups. It was 1.76 days in group A patients and 1.67 days in group B patients.

Bile culture was obtained in 34 patients. In 15 patients (44.12%) bile culture was positive. Six patients were in group B and nine patients were in group A.

Bile culture results were not different between groups (p=0.464). Drain was removed on postoperative day 1 in all patients. None of the patients had wound infection at the time of discharge and after 1 week.


LC has become the standard of care for symptomatic gallstones. At present, biliary complication after LC in patients with uncomplicated cholelithiasis is 0.2%–0.9%.3 4 Drains should be avoided as it does not reduce complications or detect them early after LC in biliary colic.1

LC in ACC is difficult due to inflammation in the Calot’s triangle and pericholecystic areas. It is not easy to identify structures in Calot’s triangle in ACC due to inflammatory adhesions and chances of vascular and biliary injuries may be increased.5 However, large-scale studies have shown that biliary and bleeding complications are similar to biliary colic.6 7 Many surgeons still prefer to drain after LC in ACC. There have been few studies assessing the role of drains in ACC undergoing LC.

In our study, we evaluated the role of drain in patients with ACC undergoing LC. The following parameters were noted: postoperative requirement of imaging and guided aspiration, necessity of continuing antibiotics, rate of wound infection, postoperative pain and duration of hospital stay.

There was no difference between two groups with respect to postoperative imaging or image-guided aspiration (p=0.311). Both groups did not differ when considering the doses of antibiotics received (p=0.238). The pain scores at 6 hours (p=0.006) and 24 hours (p=0.003) following surgery were less in group B than in group A. There was no difference in both groups with respect to wound infection and duration of stay. We did a subgroup analysis on bile culture positivity which was not different in both groups (p=0.57).

Drains are placed to detect leakage of blood or bile early or prevent any collection postoperatively. But multiple studies have disproved this hypothesis. In a review on this topic (n=8423), there was a subhepatic collection or abscess in up to 7% of patients with drain and up to 4% of patients without drain.8 Drain may be blocked by omentum within 48 hours.9 Hence, absence of blood or bile through drain should not be interpreted as absence of postoperative bleeding or bile leakage. One randomised controlled trial (RCT) reported one case of postoperative haematoma in a patient with drain. This patient did not have any bloody discharge through drain postoperatively. It was suspected clinically afterwards and imaging revealed collection. This is considered to be a limitation of drain.6

In our study, only one patient who had continuous pain and tachycardia postoperatively in ‘no drain’ group required USG, which did not reveal any collection. This patient improved with antibiotics and was discharged on day 4. In all the patients with drain in our study, the drain effluent was 50 mL or less on postoperative day 1 and the tube was removed within 24 hours.

In a retrospective study (n=103), only one patient who had drain developed intra-abdominal collection postoperatively which was aspirated radiologically.10 An RCT has reported 1% (two patients out of 193) incidence of intra-abdominal collection (one patient was in drain group).9 The results of these studies also question the need for a drain.

Fluid collection is a common postoperative finding, and most fluid collections are asymptomatic and absorbed by the peritoneum. A pilot randomised study has shown that all subhepatic fluid collections in USG on the seventh postoperative day disappeared by fourth week.11 Ultrasonographic studies clearly demonstrated that most postcholecystectomy collections remained asymptomatic and were absorbed.12 13

Postoperative intra-abdominal collection can lead to abscess formation and bile spillage can increase chance of wound infection. The main reason of using drain after LC is to prevent intra-abdominal collections. But different studies have shown that intra-abdominal collections are more in patients with drain.1 5 This may be due to the foreign body reaction caused by the drain. Even though the role of the drain is to remove fluid and debris, studies have shown that the drain itself can act as a portal of bacterial entry into peritoneal cavity and cause intra-abdominal or drain site infection. In our study, none of the patients clinically had features of intra-abdominal collection. Only one patient was suspected to have collection but imaging showed no collection. None of the patients in the study group developed wound infection. It is stated that drain gives a false sense of security.

One of the objectives of our study was to assess the duration of antibiotics in ACC. When we started this research there were no established guidelines on perioperative antibiotic therapy. Antibiotics were given regularly in ACC for several days postoperatively without any scientific basis. Antibiotic treatment, in general, is associated with adverse effects, greater costs and development of bacterial drug resistance. One RCT has confirmed that the disease course is not affected by antibiotic therapy in mild ACC.13

In our study, only eight out of 42 patients required more than three doses of antibiotics (three patients for 3 days and five patients for 4 days) . All of them had moderate ACC. These are patients with gangrene or empyema of GB which made us continue postoperative antibiotics. None of the patients required change in antibiotics in the course of the disease.14 Tokyo Guidelines 2018 recommend that in cases of mild and moderate ACC, antibiotics can be stopped within 24 hours after cholecystectomy, and in gangrenous or emphysematous cholecystitis or in perforation, antibiotics should be continued for 4–7 days.14

Our subgroup analysis of bile culture taken during surgery showed that 19 samples did not show any growth of organisms. Fifteen patients in this group had mild acute cholecystitis. This shows that mild ACC may be more of an inflammatory process rather than infective process. Our study results are consistent with the recently published RCT on extended versus single-dose antibiotic prophylaxis in mild ACC where they found that there is no difference in infectious complications in both groups.15 So it is justified that a single dose is adequate for patients with mild ACC undergoing LC .

Drains cause postoperative pain through the irritation of subcutaneous nerves or by direct stretching of the peritoneum at the insertion site. Few studies have used visual analogue scale to assess pain scores. Those studies have shown increased pain in patients with drain.5 16 We have used NRS for pain as it has the advantage of being self-administered. We found that patients with drain have more pain compared with the no drain group. 47.62% of patients in group A had NRS of 5 compared with 14.29% in group B at 6 hours. Similarly at 24 hours the maximum NRS was 4; 47.62% in group A and 9.52% in group B. Pain score at 6 hours in our study was less than compared with a study conducted by Kim et al.6 This may be explained by the intraperitoneal instillation of bupivacaine done in our study.

We calculated the duration of stay from the day of surgery. Sixteen patients were discharged on day 1, all were mild ACC. Nineteen patients were discharged on day 2. Seven patients were discharged on day 3 out of which four were moderate ACC. On analysis there was no difference between both groups (p=0.51). This result was similar to the previous two RCTs done on the same topic.6 10 It is understood from our study that drain may not prolong the duration of stay, whereas it is the severity of the disease which prolongs the stay.

Early LC is recommended by a meta-analysis published in 2015 and confirmed by other studies.17 But the definition of early surgery differed in studies varying from 3 days to 1 week. Tokyo Guidelines published in 2018 stated that early LC should be advised in patients with mild and moderate ACC. They defined early as 72 hours to 1 week .14 But majority of the surgeons still follow a delayed approach in moderate severe ACC fearing complications.

In our study, 34 out of 42 patients were diagnosed with mild ACC and they all underwent early LC (within 72 hours). We had eight patients with moderate ACC, all underwent surgery within 1 week of admission. In our study, none of the patients had postoperative complications like bleeding or bile leakage. One patient in moderate severe ACC had cystic duct torn from cystic duct-common bile duct junction which was suture repaired. This patient was excluded from the study as the surgeon himself has placed a drain. Since the injury was detected and promptly managed, the patient did not have any postoperative bile leakage. So it is clear that even in moderate severe ACC, early LC can be safely done, provided the surgeon is well experienced. Because the incidence of these complications is rare, routine drainage is not desirable.

As mentioned earlier, we have drained in special situations where bile leak or bleeding is expected. These patients were excluded from the study. We justify the placement of drain in these situations because the bile is irritant to peritoneal cavity and can cause biliary peritonitis.


This study compared the role of drain in patients with ACC undergoing LC. We assessed six objectives to come to the conclusion.

Pain was more in patients with drain compared with the no drain group. There was no difference in the requirement of postoperative imaging between both groups. Patients with mild ACC required only three doses of antibiotics and more doses were required in patients with moderate ACC. Wound infection rates and duration of hospital stay were not different in both groups.

In mild and moderate ACC, drain insertion causes more pain, and except in cases where bleeding or bile leak is anticipated, it should be avoided.

Main messages

  • Drain placement in laparoscopic cholecystectomy for acute calculous cholecystitis is controversial, unlike biliary colic where it is not recommended.

  • This randomised controlled trial concludes drain placement does not detect complications.

  • Drain placement increases postoperative pain.Current research questions Is drain placement necessary in laparoscopic cholecystectomy?What is already known on the subjectDrain is not required in laparoscopic cholecystectomy in biliary colic.



  • Contributors SGu and SGh conceived and designed the study. MVV collected the data, performed the analysis and wrote the manuscript. MC, AM and SDB contributed and analysed the data. SGu and SGh reviewed and edited the manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval Institutional ethics committee PROT/DNB/31/GIS/03/2016 Dr 31 May 2016.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available upon reasonable request.