Abstract
Abstract
Background:
Symptomatic gall stone disease requires early emergency treatment to prevent complications. This early treatment is often delayed due to difficulty in the diagnosis and management of concomitant choledocholithiasis. Intervention with preoperative endoscopic retrograde cholangiopancreatography (ERCP) is associated with complications and known to be unnecessary in most cases. We follow a strategy of providing early cholecystectomy with selective utility of antegrade stent in cases of choledocholithiasis. Our main aim is to present our technique and results.
Method:
We conducted a 3-year (January 2014 to January 2017) review of a prospectively maintained database of our practice of performing routine intraoperative cholangiogram (IOC) during laparoscopic cholecystectomy (LC) and when choledocholithiasis is encountered on IOC; a transcystic antegrade biliary stent is inserted to decompress the common bile ducts (CBD) and facilitate postoperative ERCP at later date.
Results:
Of the 411 cholecystectomies performed, 77.3% were females with mean age of 48 years. Seventy-four patients were found to have CBD stones (CBDS) on IOC. Antegrade stents were successfully deployed in 69 cases. Even though Antegrade stents were done more frequently in emergency admissions (P = .001); this did not increase the length of hospital stay (LOHS) (P = .752) or the rate of complications (P = .171). However, doing a preoperative ERCP significantly increased LOHS (P = .001), and 67% of these needed two or more ERCP for complete clearance of CBD and had more complications. Nine (15.2%) out of 59 patients with pancreatitis had CBDS on IOC and were successfully managed with antegrade stent.
Conclusion:
This strategy can be followed by general surgeons, enabling them to perform LC in the presence of choledocholithiasis during acute admissions including pancreatitis. It does not require any specialist skills in CBD exploration and also eliminates unnecessary preoperative ERCP and avoids its potential complications.
Introduction
Gall stone disease continues to be an enormous health care problem accounting for a high number of emergency admissions. 1 There is growing evidence from various studies and systematic reviews supporting emergency laparoscopic cholecystectomy (ELC) for symptomatic gall stone disease including acute gallstone pancreatitis during the same index admission as soon as patients are stable.2–7 This has been shown to reduce recurrent attacks of pain, cholecystitis, cholangitis, and pancreatitis as well as repeated hospital admissions, benefiting patients and making it more cost effective. 8
One of the reasons for a delay in laparoscopic cholecystectomy (LC) during the acute admission is accurate preoperative diagnosis of concomitant common bile duct stones (CBDS), which has an incidence of 10% to 20%.9–11 One of the most popular strategies of managing cholelithiasis when concomitant CBDS is suspected is a two-stage management, preoperative endoscopic retrograde cholangiopancreatography (ERCP) followed by LC. The problem with this approach is that it causes delay in LC, is more cost intensive, causes longer hospital stay and sometimes can cause further complications related to ERCP.10,12,13 Additionally, the preoperative ERCP can be negative (i.e., no CBDS detected at ERCP in spite of radiological and biochemical corroboration) 10 and also has a low therapeutic rate (30%) 10 and a high false negative rate (missed stones) of 8% to 33%.10,14–16 This makes preoperative ERCP unnecessary in most of cases and also exposes patients to unnecessary postprocedural complications. The high false negative rate causes false reassurance, only for these patients to present at a later date post LC (without intraoperative cholangiogram [IOC]) with complications of a bile leak, jaundice, cholangitis, and pancreatitis.10,12–17
To overcome this, laparoscopic common bile duct exploration (LCBE) during LC was promoted as a single-stage management of concurrent cholelithiasis and choledocholithiasis, but this is dependent on availability of advance facilities and specific surgical expertise. Also this approach has not been proven superior to a two-stage management (i.e., combining LC with ERCP). The large systematic review, meta-analysis, and randomized control trials conclude that there is no advantage in terms of failure rates, retained stones, complication, mortality, and morbidity.13,18–20 The other single-stage technique such as laparoscopic endoscopic rendezvous techniques in managing concomitant CBDS during LC has failed to gain popularity, because it needs more equipment, extra specialists, and prolongs surgery time, thereby causing burden on the day case LC list. Hence, very few centers practice laparoscopic endoscopic rendezvous techniques, so two-stage management of gallstones is still very popular among surgeons.10,21–24
Our strategy in all patients (with acceptable anesthetic risk) with symptomatic gall stone disease including acute cholecystitis with normal or improving liver function test or stable patients with gallstone pancreatitis is to undergo ELC combined with IOC. In the case of no CBDS on IOC, then an ELC is completed, thereby reducing the need for either preoperative magnetic resonance cholangiopancreatography (MRCP) or ERCP. If the IOC detects CBDS we deploy an antegrade biliary stent to decompress the CBD, and we arrange an elective day case ERCP at later date to retrieve the stent and CBDS.
These antegrade biliary stents have been used since the 1990s to decompress CBD following LCBE for choledocholithiasis mainly to avoid the use of T-tube.8,25–29 Martin et al. 2002 reported its usefulness in transcystic antegrade biliary stenting in case of CBDS encountered during LC on IOC, without the need for CBD exploration. However, this required post-LC ERCP to remove the stones and stent. 30 There have been no subsequent published reports regarding antegrade stents being used in this particular way.
Our main aim is to present our technique and results in performing LC and antegrade CBD stenting in patients with concomitant CBDS. Our secondary aim is to see if this reduces the number of ERCPs and its cost effectiveness.
Methods
We conducted a 3-year (January 2014 to January 2017) review of a prospectively maintained database of a single hospital and surgeon's practice of performing routine IOC during LC and when choledocholithiasis is encountered on IOC; a transcystic antegrade biliary stent is inserted to decompress the CBD and facilitate postoperative ERCP at later date. Only patients with gallstone disease were included. We excluded patients who were managed by other surgeons in a traditional way—that is, had preoperative ERCP followed by referral for LC–or in whom IOC was contraindicated, such pregnant women, or in case of allergies. Patients who had subtotal cholecystectomies and/or for whom the IOC was not technically feasible were also excluded from this study.
Technique
Disection and preperation
We perform LC using the four port technique. Pneumoperitoneum is created by inserting a 10 mm infraumblical port. A 10 mm epigastric and two 5 mm right upper quadrant ports (anterior axillary and mid axillary line) are inserted under vision. Calot's triangle is dissected safely, ensuring a posterior critical window of safety delineating the cystic duct (CD) and artery.
Intraoperative cholangiogram
We perform an IOC using an Olsen endoscopic cholangiography set (Cook Medical, REF G05927), which comes with a percutaneous cannula and needle and a cholangiogram catheter (CC) (4.0F/43 cm). A 2 mm stab wound is made in the right subcostal region with a knife between the two 5 mm ports, and the cannula and needle is inserted under vision toward the gall bladder. Once inside, the needle is withdrawn and the CC is flushed and connected with a 10 mL syringe with normal saline, ensuring no air bubbles. The CC is inserted via the cannula so that the tip easily reaches the CD.
Next, a titanium clip (Teleflex medical ref-003200) is applied to the CD just below the Hartmann's pouch. The CD is partially cut with scissors, just enough to allow the tip of CC. Any suspicious debris or stone in CD is milked up and removed via the cut. Then about a centimeter of CC is inserted into the CD, secured in place using a clip on CD over the CC, and flushed with saline to ensure free flow and no leaks. Then an IOC is done using radiopaque dye (Iopamidol, Nefopam 300) and an X-ray image intensifier.
If there are no CBDS and anatomy is well defined, we remove the CC and clip the CD and artery before dividing them and proceed to completing LC.
Transcystic antegrade stenting
When CBDS are detected on IOC, we do a transcystic antegrade biliary stenting. No attempt is made to clear the CBDS. On the basis of a positive IOC, an antegrade stent is inserted using the Seldinger technique. The CC is removed and an open-end flex-tip ureteral catheter (7F, 70 cm; Cook Medical, REF G14809) is inserted via the cannula into the CD. Then a guidewire (Guide wire L, flex tip 3 cm, length 180 cm, 0.98 mm diameter; Terumo Europe Corp.) is inserted via the ureteral catheter into the CBD (Fig. 1A). This is advanced into the duodenum under X-ray guidance. Once the guidewire is in satisfactory position the ureteral catheter is withdrawn keeping guidewire in position. A double pig tail end biliary stent (Zimmon Biliary stent, REF G22161, 7F, 4 cm; COOK Medical) is threaded over the guidewire and eased into the CBD via the CD under image guidance (Fig. 1B). A pushing catheter (Cook medical, 7F, 170 cm, G21774) is cut to size (40 cm) and used to place the stent in satisfactory position (Fig. 1C). As soon as the stent disappears from view and the distal end is seen at the ampulla on imaging, the guidewire is withdrawn, keeping the pushing catheter in place to prevent stent displacement. As the guidewire is pulled out, both ends of stent curl up traversing across the ampulla (Fig. 1D). The final stent position is confirmed by image guidance. Then the cholecystectomy is completed. We routinely do not use drains after the procedure.

Postoperative ERCP
We arrange a postoperative day case ERCP for all patients in whom an antegrade biliary stent was inserted to decompress the CBD in case of CBDS within 12 weeks. This time enables patients to recover fully from the LC and also makes the CBD stones easier to extract.
Statistical analysis
Descriptive statics were calculated for all variables. These included mean and standard deviation (SD) or median and interquartile range for continuous factors and percentages % for categorical factors. Categorical variables were assessed using Pearson's chi- squared test (χ2) or fishers exact test as appropriate to for significance of difference between groups. The quantitative and ordinal variables were tested using Kruskal-wallis test, Wilcoxon rank sum test or paired comparisons as appropriate. Statistical analysis was done using 2013 IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.
Results
We performed 411 LC with an IOC over 3 years, among which there were 93 (22.6%) male and 318 (77.3%) female patients. The mean age was 48 years (16 to 87 years). Only patients presenting with symptomatic gall stone disease were offered LC. Of these, 211 were emergency admissions and underwent LC during their index admission. The main diagnosis at presentation was cholecystitis (253; 61.5%) (at least 6 [1.4%] were perforated gall bladder); followed by biliary colic (69; 16.7%); pancreatitis (59; 14.3%); obstructive jaundice (18; 4.3%); cholangitis (5; 1.2%); and Mirizzi syndrome (2; 0.48%).
A majority of patients were anesthetic risk assessment (ASA) grade-2 (224; 54.5%), followed by ASA-1 (143; 34.7%), ASA-3 (40; 9.7%), and ASA-4 (4; 0.97%). Our overall conversion rate from laparoscopic to open cholecystectomy was 3.6% (15).
Preoperative imaging
Ultrasound scan showed dilated CBD (range 8 mm to 1.8 cm) with suspicion of CBDS in 80 patients (19.4%). Forty-one (51.2%) of these 80 patients had CBDS confirmed on IOC. Thirty-nine patients had a false positive ultrasound scan, as no CBDS found on IOC. Twenty of 331 patients with normal US scans had CBDS on IOC, giving a false negative rate of 10.2%. The sensitivity, specificity, positive predictive value, negative predictive value (95% confidence interval [95% CI]), and accuracy of US scan was 67.2%, 88.8%, 51% (CI: 40.4–61.8), 93% (CI: 90.8–96), and 85.6%, respectively.
MRCP was done in selected patients who had an inconclusive ultrasound scan but abnormal liver function test (raised alkaline phosphatase, alanine aminotransferase or bilirubin) or had suspicion of underlying malignancy or CBD stricture. Out of 55 patients who had an MRCP scan, 13 patients were confirmed to have CBDS. All the 13 had successful IOC and CBDS were present in only 9 of these patients. Eight of these were treated with antegrade biliary stents and the remaining one had Mirizzi syndrome and underwent open CBD exploration. Also, 4 patients with normal MRCP had CBDS on IOC. The sensitivity, specificity, positive predictive value, negative predictive value (95% CI), and accuracy of 69.2%, 90.4%, 69.2% (CI: 42.3–87.3), 90.4% (CI: 77.9–96.2), and 85.4%, respectively.
Pancreatitis
Of the patients who presented with mild to moderate pancreatitis, 59 patients underwent LC during their index admission. All 59 patients had successful IOC, and CBDS were confirmed only in 9 (15.2%). All these 9 patients with CBDS on IOC had CBD decompressed with antegrade biliary stent during LC. The average postoperative length of hospital stay (LOHS) was 4 days. There was no increased risk of complications in this group compared with non-pancreatitis LC (P = .02).
IOC and antegrade stent
Of the 411 patients who had successful IOC during LC, 74 of these were found to have CBDS. Antegrade stents were successfully deployed in 69 cases. Three patients in whom the stent failed to deploy underwent immediate postoperative ERCP and remaining 2 had Mirizzi syndrome, and they underwent open exploration. Average age of patients undergoing Antegrade stent is 48 ± 18 (mean ±2SD). The female to males ratio was 54:15. The majority of patients undergoing antegrade stent presented with cholecystitis (39), followed by obstructive jaundice (10), pancreatitis (9), biliary colic (9), and cholangitis (2).
Of the 74 patients who were diagnosed with CBDS, 20 had a normal US scan, 4 had normal MRCP, and 30 had normal bloods. Incidental CBDS with both normal imaging and bloods were found in 25 patients.
The patients who had IOC and antegrade stent were similar in demographic profile compared to patients who only needed LC (as no CBDS on IOC) (Table 1). They were similar in terms of age (48 years P = .9) and gender distribution (P = .285), and the co-morbidities/ASA grades were similar in both groups (P = .724). Even though the antegrade stents were done more frequently in emergency admissions/hot gall bladders compared with elective admissions (P = .001); doing antegrade biliary stents did not increase the LOHS (P = .752) or the rate of complications (P = .171).
Comparing Patients Who Had CBDS on IOC Treated with LC and Antegrade Biliary Stent versus Patients Who Had No CBDS on IOC were Only Treated with LC
ASA, anesthetic risk assessment; CBDS, common bile duct stones; IOC, intraoperative cholangiogram; LC, laparoscopic cholecystectomy.
However doing a preoperative ERCP followed by LC significantly increased LOHS (P = .001). Patients who had two or more ERCP's also had cumulative prolonged LOHS.
The average time taken to perform IOC was 5 minutes. When CBDS were encountered, inserting an antegrade biliary stent added an extra 5 minutes to the LC procedure.
Postoperative ERCP
In 66 patients with antegrade stents inserted, CBD clearance was achieved with only one postoperative ERCP. However, the remaining 3 patients with antegrade stents needed two ERCPs to gain full clearance of CBD and removal of stent.
This compared with a contemporaneous group of patients treated in a traditional way for CBDS (n = 79) either with preoperative ERCP or LC and postoperative ERCP, almost 67% of these needed two or more ERCP for complete clearance of CBD. The relative risk of 5.11 (CI: 1.5–17.2) (P = .0084), thereby number needed to treat of 5.595 (95% CI: 15.02–3.437). In patients who were treated with preoperative ERCP, the risk of complications increased with number of ERCPs needed to clear the CBDS (P = .001) (Table 2).
Total Number of Endoscopic Retrograde Cholangiopancreatography in Antegrade Stenting Group versus Traditional Group
ERCP, endoscopic retrograde cholangiopancreatography.
Complications
The overall rate of 30-day complications post LC was 6.5% (27 patients); these include bile leak (4), bleeding (3), collections in GB fossa (10), surgical site infections (6), chest infections (3), and PE (1) (Table 3).
Overall Complication Rates of All Patients Included in the Study, None of Which Occurred in Antegrade Stent Group
The 4 patients with bile leak were managed with drains (3 ultrasound guided and 1 surgical) and postoperative ERCP. All resolved with this management. The mean hospital stay was 6 days. None of these occurred in antegrade stent group.
Ten patients developed deep collections, in either GB fossa or subphrenic space. All of them were treated with antibiotics, and 6 of these required radiological guided drains. All of these patients resolved without further surgical intervention.
Costs of doing an antegrade stent adds £108 to the LC including instruments and radiograph charges, where as an ERCP with stent costs £1302.
Discussion
The challenge in managing CBDS is that accurate preoperative diagnosis is difficult and a significant number (up to 20%) are detected when routine IOC is undertaken during LC. Studies have found that preoperative imaging (US/MRCP), biochemistry test (LFT), presence of gallbladder or CD stones, the time interval between ERCP and LC and performance of sphincterotomy, mode of presentation, and pancreatitis are poor predictors of which patients will harbor residual CBD stones.10,14,15
In our study, preoperative ultrasound, although good in detecting cholelithiasis, has low sensitivity (67.2%) in detecting choledocholithiasis. These findings are similar to a recent Cochrane review showing that sensitivity and specificity of US is 73%. 32 MRCP, which has a higher sensitivity and specificity of 93% and 96%, respectively, 33 it is not widely available and has a long waiting list in a health care system like NHS. The other challenge CBDS pose is that in spite of a normal liver function test and preoperative imaging showing normal CBD, CBDS can be detected on routine IOC. The converse is also true–that is, the CBDS may be present on preoperative imaging and absent on IOC during LC. Twenty of 331 patients in our study with normal US scans had CBDS on IOC, giving a false negative rate of 10.2%. This may be because the gallstones/CBDS are mobile and may move from presentation to intervention (LC ± ERCP) through biliary ducts, giving false positive and false negative results on preoperative imaging. These factors make the preoperative diagnosis of CBDS difficult. Consequently the preoperative ERCP may be negative or CBDS are detected on routine IOC during LC in spite of negative diagnostic imaging and also sometimes in case of negative preoperative ERCP.10,15,16
With the evolution of LC, simultaneous LCBE was promoted as a single stage, safe, and cost-effective technique.8,20,34,35 However its general adoption by majority surgeons has been very limited, since most practice the two-stage technique of LC combined with ERCP,10,21–24 probably because LCBE is resource intensive, time-consuming, requires specialist skills, and is not free of complications,10,21,22,36 with similar rates of failure, retained stones, complication, mortality, and morbidity.13,18,19,20 Also, advancements in ERCP skills and techniques make LBDE less attractive. As long as the CBD is decompressed (with a stent), clearance of the bile duct at the time of surgery less is imperative and a postoperative ERCP can be safely performed at a later date.
It is now well established that patients who present with symptomatic gall stone disease, acute cholecystitis, or even mild pancreatitis will benefit from having LC and CBD decompression during the index admission.3,4–8 The main problem with two-stage management is that patients might undergoing unnecessary preoperative ERCPs, which could lead to complications and delay LC.10,12,13 The reason for delay is that the facilities, availability of equipment, and expertise in ERCP are not always available, especially out of hours, so there is usually a long wait for preoperative ERCP and delayed LC during acute admission. Therefore we practice a policy of doing LC during acute admissions and avoid delay while awaiting preoperative ERCP in suitable patients.
One significant finding from our study is that patients who underwent preoperative ERCP subsequently needed future ERCP to achieve clearance of CBDS. We also found that in 22.2% of preoperative ERCPs, no CBDS was found in spite of CBDS confirmed on preoperative imaging. Similarly, other studies have shown that the prediction of CBDS is difficult, and this has lead to a high number of unnecessary preoperative ERCPs (∼56%–68%) or putting patients at risk of a high level of complications due to repeated ERCPs, including pancreatitis (2%–11%), perforation (1%–4%) or bleeding (2%–4%).9,11,12,14,14,15,17 ERCP diagnostic success rate is 76%–97%, with an inability to cannulate of 5%. Taylor et al. proposed selective preoperative ERCPs to reduce the rate of unnecessary ERCP by selecting appropriate patients. 16 Serum bilirubin was found to be the best predictor of choledocholithiasis. 16 Coté et al. 14 found the incidence of CBDS was 8% in patients who had negative preoperative ERCP. He also noted that in cases where routine IOC were not performed, they risk false reassurance and risk representing with pancreatitis, jaundice, and cholangitis. Similarly, Pierce et al. 15 found that the incidence of residual CBDS is 16.9% in patients who underwent preoperative ERCP and 8.3% in patients with negative preoperative ERCP, yet another study found that preoperative ERCP positively confirmed CBDS only in 44% of patients who were referred for preoperative ERCP after imaging showed dilated CBD and CBDS with cholestatic liver enzymes on blood test. In this study the therapeutic rate of preoperative ERCP is 30%. 10 Another study has found 33% incidence of residual CBDS after presumed successful preoperative ERCP. 16
Postoperative ERCP has been very reliable in clearing the CBDS, and most of the patients (95.6%) in our series did not require more than one ERCP. The endoscopist reported that having a stent in place makes the postoperative ERCP easier. It helps to identify the ampulla and complete a sphincterotomy. It also reduces the risk of pancreatitis, perforation, and bleeding, as the endoscopist can follow the stent to CBD and avoid cannulating the pancreatic duct. 10
So far they have been no consensus on performing routine IOC, but the advantages in doing them are clear that is, they help detect missed CBDS missed, delineate anatomy and have a role in preventing bile duct injury. 37 As we do IOC in all patients unless contraindicated it on an average adds about 5 minutes to operating time, as the whole theater team is familiar with the routine procedure. Surgeons who do IOC selectively take much longer and have unfamiliarity in reporting IOC. 38 The mean time taken to perform an IOC was 5 minutes (range 4.3 to 18 minutes), and our rate of successful IOC was 92%. In keeping with published literature, the majority of studies cited success rates over 90% (range from 82% to 97%).39–43
Similar to us, others have also found that antegrade biliary stents save time, are easy to do, and effectively decompress CBD in presence of CBDS, 10 whereas yet others have used it to decompress CBDS after LCBDE. 44 Some studies have also used antegrade stent after CBD exploration to avoid T-tube and have found it to be useful and to reduce complication rates. 26 Some evidence suggests that another advantage of inserting stents is that insertion of a stent makes is easier to later retrieve CBDS that are irretrievable at ERCP.45,46 This may also be the reason why in our series, 95.6% (66 out of 69) patients with antegrade biliary stents had only one postoperative ERCP, whereas 67% of the concurrent patients group who were managed in a traditional way with preoperative ERCP followed by LC needed two or more ERCP.
Patients with mild to moderately severe pancreatitis can also be offered this management strategy, as most patients with pancreatitis do not have a CBDS on IOC.3,19,33,36 In our study, CBDS were present in only 15.2% of patients with pancreatitis. We also found that doing antegrade biliary stent in patients with acute pancreatitis did not increase complications or their LOHS.
When done routinely, an antegrade biliary stent is easy to do and takes an average of 5 minutes. We also had a good success rate (95.8%) in inserting an antegrade biliary stent when CBDS were found on IOC. However, if the stent fails, the options are either an open CBD exploration or a postoperative ERCP.
Our data shows that doing routine IOC and antegrade biliary stents in case of CBDS is safe and expedites LC. It also doesn't increase complications or LOHS. This is one of the good alternative techniques for managing CBDS during emergency admissions of symptomatic gallstone disease and offering LC without delay awaiting MRCP or ERCP.
This strategy also reduces the number of ERCPs required to clear CBDS, thereby reducing the rate of complications, morbidity, and the resources required to do repeated ERCPs. As it requires fewer resources and is cheaper, it can be widely applicable without advance LCBDE or the need for preoperative ERCP. As it is cheaper than ERCP and reduces the LOHS, it is more cost effective. Also, the MRCP can be done more selectively when an ultrasound scan is inconclusive.
The cost of doing an antegrade stent adds £108 to the LC, including instruments and radiograph charges, whereas an ERCP with stent costs £1302.05.
This study does have some limitations. Our intention here is mainly to describe the technique and its feasibility. We do acknowledge that this is not an RCT or comparative study, and further prospective studies are needed to see whether routine IOC can reduce number of requests for MRCP and ERCP. Further studies are also needed to compare this strategy with the traditional techniques. In our hands we find that this strategy is effective, less resource intensive, reduces number of hospital stays, and is easy to teach and learn.
Conclusion
Although there are several approaches currently employed to treat CBDS, they all have limitations. We feel that removal of the gall bladder along with gallstones and performing IOC ensuring that there are no stones left in CBDS is possibly the only way to be sure that no CBDS are left behind, which would cause problems later. LC with an IOC and antegrade stent (in case of CBDS) in selected cases is a safe and cost-effective option. It is a safe and reliable procedure in surgeons experienced in minimal access surgery. It also prevents any delay in surgical treatment while awaiting an ERCP and avoids unnecessary preoperative ERCPs, missed stones (false-negative ERCP), and also reduces the number of ERCPs required to clear CBDS.
Footnotes
Disclosure Statement
No competing financial interests exist.
