Gallstone pancreatitis: does discharge and readmission for cholecystectomy affect outcome?

Abstract

Background

Conventional surgical wisdom is that a patient with gallstone pancreatitis should have the gallbladder removed during their initial hospitalization. However, patients are now often discharged to await operating room availability.

Methods

A retrospective review of all cases of gallstone pancreatitis at the Foothills Hospital between 1992 and 1996 was undertaken. Patients with a first attack of mild gallstone pancreatitis were studied.

Results

In all, 164 patients were identified: 90 patients were discharged for readmission cholecystectomy (discharged group), and 74 patients had the cholecystectomy before discharge (in-hospital group). Over the 5-year time period the proportion of patients discharged for readmission cholecystectomy increased from 27% to 67% (p<0.01). The total number of days waited for operation was greater in the discharged group versus in-hospital group: 40±69 days versus 8±10 days respectively (mean±SD). There was a trend towards an increased total number of days in hospital in the in-hospital group, 15.5±17 days versus 10.7±16 days. In the discharged group 20% (18 of 90) of patients experienced an adverse event requiring readmission while awaiting operation. Three had documented recurrent pancreatitis, 10 experienced recurrent pain, and 5 developed acute cholecystitis. There were no deaths in either group.

Discussion

Twenty percent of patients with gallstone pancreatitis who are discharged to await operating room time (average wait 40 days) will require readmission for biliary symptoms.

Introduction

It is well established that cholecystectomy results in a decrease in the relative risk of acute pancreatitis in patients with known gallstone, to a risk comparable to that of the general population ¹. However, there has been controversy as to the proper timing of operation for patients with acute gallstone pancreatitis. Much of the debate has focused on the timing of operation in the acute situation for patients with a first attack of gallstone pancreatitis. While recommendations differ depending on the clinical situation, some advocate immediate operation ²,³,⁴, while others recommend that the initial attack be allowed to settle before cholecystectomy ⁵,⁶,⁷. Another part of the controversy centers on the timing once the acute attack has settled, with some advocating discharge home and elective readmission for cholecystectomy in months to weeks while most now recommend cholecystectomy as soon as possible during the initial hospital admission ⁵,⁶,⁷,⁸,⁹,¹⁰,¹¹,¹²,¹³,¹⁴,¹⁵,¹⁶,¹⁷,¹⁸,¹⁹,²⁰,²¹,²²,²³,²⁴.

In the early 1990s a gradual, although unofficial, change of practice occurred in the tertiary care system described. Patients admitted with a first attack of acute gallstone pancreatitis were waiting long periods in hospital for operation, presumably due to a paucity of operating room time. Previously, most patients had operation during their initial admission to prevent recurrent pancreatitis, biliary symptoms, and other complications. Owing to these long hospital waits surgeons began discharging their patients for readmission cholecystectomy in the near future. The following study was undertaken to determine the natural history of patients admitted to one institution with a first attack of acute gallstone pancreatitis during the 5-year period when this transition in care occurred.

Patients and methods

A retrospective chart review was undertaken of all patients admitted to the Foothills Hospital, a tertiary care hospital in Calgary, Canada, with a first attack of acute gallstone pancreatitis, using ICD-9 codes for the years 1992–1996. Information collected included basic demographic data, Ranson's criteria, date of diagnosis, date of operation, whether or not the patient was discharged before operation, total number of admissions and days in hospital, operation performed, and operative complications. Inclusion and exclusion criteria are listed in Table 1. Only patients with a limited attack of acute gallstone pancreatitis were included. The hospital records of the two other major hospitals in the city were also reviewed to rule out the possibility that patients were lost to follow-up.

Table 1. Inclusion and exclusion criteria.

Inclusion criteria:

1. Admitted with a first attack of gallstone pancreatitis

2. Lipase elevation >400 U/L

3. Radiographic confirmation (gallstones on ultrasound, CT or ERCP)

4. Cholelithiasis reported at cholecystectomy and pathology

Exclusion criteria:

1. ICU stay

2. Urgent laparotomy for acute abdomen

3. Prolonged pain (>3 days), pseudocyst formation, necrosis on CT scan

The patients were divided into two groups based on whether or not they were discharged before cholecystectomy. Individual surgeons made the decision to discharge, based on the availability of operating room time. The two groups were compared with respect to waiting time to operation, total number of days in hospital, rates of discharge, complication rates, and number of admissions. Statistical tests used were a χ² analysis for discrete variables and a two-tailed t test for continuous variables. Results are expressed as means±standard deviation.

Results

A total of 164 patients met the inclusion criteria for the study. Ninety of these patients were discharged before elective readmission for operation (discharged group) and 74 were kept in hospital for definitive operation (in-hospital group). The groups were comparable with respect to age, male:female ratio, ASA status, and admission laboratory values (Table 2).

Table 2. Demographic and clinical data in discharged versus in-hospital groups.

Variables	Discharged patients (%)	In-hospital patients (%)
Number of patients	90	74
Age (years)	51.2	55.4
Male:female	1:2	1:1.9
ASA status 1	75 (83)	53 (72)
2	11 (14)	15 (20)
3	4 (4)	6 (8)
Mean lipase (U/L)	6043	6414
Mean WBC (×10/9 L)	11.3	12.7

There was a trend towards an increasing proportion of patients discharged before cholecystectomy over the 5-year period, as follows: 27% (8/29) in 1992, 50% (13/26) in 1993, 69% (20/29) in 1994, 54% (19/35) in 1995, and 67% (30/45) in 1996.

As shown in Table 3, there was a greater number of days waited for operation in the discharged group than the in-hospital group. There was also a trend for the in-hospital group to spend a greater number of days in hospital than the discharged group (p = 0.07).

Table 3. Outcome variables in discharged versus in-hospital groups.

Variable	Discharged (%)	In-hospital (%)
Total number of days in hospital	10.7±16	15.5±17
Readmissions/adverse events (pre/postop)*†	18/3	0/4
Length of wait to operation (days)†	40±69	8±10
Number of ERCPs	57 (63)	46 (62)
Conversion rate	9/74 (12)	8/90 (9)
IOC rate	39 (43)	41 (55)

ERCP, endoscopic retrograde cholangiopancreatography; IOC, intra-operative complication.

*Adverse events requiring preoperative readmission included acute pancreatitis (3), cholecystitis (5) and recurrent pain (10).

^†p<0.001 (preoperatively).

Eighteen patients in the discharged group (20%) experienced an adverse event requiring readmission, while no recurrent symptoms or events were documented in the in-hospital group. These adverse events included repeated episodes of pain with no documented increase in lipase (n=10), recurrent pancreatitis with elevated lipase levels (n=3), and acute cholecystitis (n=5) (Table 3).

Postoperative complications including respiratory, cardiac, gastrointestinal, and wound complications were defined before chart examination ²⁵. There were no differences in postoperative complication rates, taken either as a total or when subclassified into different types of complications by system. The total number of complications was 11 (15%) in the in-hospital group and 16 (18%) in the discharged group (not significant).

Discussion

A restriction in operating room time in our hospital has resulted in a change in the management of patients who present with a first attack of acute gallstone pancreatitis. Over the study period an increasing number of patients were discharged to await elective cholecystectomy after an attack of gallstone pancreatitis. This study was undertaken to determine how these changes in health care delivery affect patients.

The results show that over the 5-year period the proportion of patients discharged to await elective operating room time increased by 40%. In addition, patients who were discharged before definitive operation for gallstones waited a significantly longer period for operation, had more emergency room visits, and had a higher rate of recurrent biliary tract symptoms (all requiring readmission) before operation than their in-hospital counterparts. These findings probably underestimate the rate of recurrent symptoms, as only patients who presented to the emergency department with readmission were included. Office visits and attacks of pain not requiring a physician visit were not captured.

A literature review has revealed many studies on the natural history of acute gallstone pancreatitis and timing of operation ²,³,⁴,⁵,⁶,⁷,⁸,⁹,¹⁰,¹¹,¹²,¹³,¹⁴,¹⁵,¹⁶,¹⁷,¹⁸,¹⁹,²⁰,²¹,²²,²³,²⁴, with widely varying waiting times and complication rates during this period. There is no definitive evidence as to the best approach to the management of a patient who presents with a first attack of acute gallstone pancreatitis, although it is accepted that for those with a mild attack, operation during the initial hospital admission is safe.

In 1995 Kennedy and colleagues ²⁶ retrospectively compared those patients booked for elective cholecystectomy with documented gallstones who had to wait for state approval requirements versus those patients who did not. They found that the patients who waited for approval had a statistically significant increase in waiting time for operation. Their morbidity in terms of number of emergency room and outpatient clinic visits when compared with those patients who did not require state authorization was also significantly increased. This study reviewed patients who had experienced symptoms attributable to gallbladder pathology, but did not specifically look at that subset of patients with gallstone pancreatitis.

As more operations are done on an outpatient basis and with increased economic pressures, the health care system in Canada has seen marked changes in clinical management. There has been little evidence so far that these changes affect patients adversely. Currently, the British Society of Gastroenterology recommends that patients with acute gallstone pancreatitis have an operation within 28 days ²⁷. Our discharged group clearly falls outside this standard with an average wait for operation of 40 days, significantly longer than the in-hospital group (8 days). The consequence of this wait was an increased risk of experiencing an event that required urgent readmission. However, waiting for operation did not appear to result in increased peri-operative morbidity.

Early operation on patients with acute pancreatitis may carry an increased risk of operative problems. Laparoscopic cholecystectomy carried out at an average of 8 days after the diagnosis of acute pancreatitis in our group of patients did not result in an increased complication rate or conversion to open cholecystectomy. This may have occurred as a result of the selection of patients with mild attacks and the practice of only operating once the pain had settled.

We conclude that the preferred management of patients admitted to hospital with a first attack of mild gallstone pancreatitis is to undergo laparoscopic cholecystectomy during their initial admission. For patients with ‘mild’ acute pancreatitis we do not recommend a waiting period once the acute attack has settled. While this study was retrospective and only examined information that could be extracted from hospital charts, there may be other advantages to following a protocol that includes first admission operation. These include psychological effects on patients awaiting operation, the increased costs of urgent readmissions, and lost productivity for those who experience recurrent symptoms that are not severe enough to come to medical attention.