World J Surg (2016) 40:849–855 DOI 10.1007/s00268-015-3337-5

ORIGINAL SCIENTIFIC REPORT

Cholecystectomy During the Weekend Increases Patients’ Length of Hospital Stay Josephine Philip Rothman1 Jacob Rosenberg1



Jakob Burcharth1 • Hans-Christian Pommergaard1



Published online: 12 November 2015 Ó Socie´te´ Internationale de Chirurgie 2015

Abstract Background A higher risk of complications and mortality has previously been proven in selected settings. The purpose of this study was to investigate whether length of stay differentiates throughout the week and register if intraand postoperative complications vary on weekends compared to weekdays. Methods The population originated from the Danish Cholecystectomy Database. It consists of adult patients, who had a cholecystectomy performed by standard four-port laparoscopic or open surgery. Adjusted analyses were used to study if day of the week had an influence on conversion, readmission within 30 days, post-operative supplemental procedures within 30 days, and variance in postoperative length of stay across the week. Results A total of 28,759 patients were included in the study. We found no difference in conversion rate, readmission within 30 days, or post-operative procedures within 30 days between week time and weekend time. A longer postoperative length of stay was observed for patients operated on Fridays and Saturdays even though surgical complication rates were alike between weekdays. Patients with acute cholecystitis had a longer length of stay on Saturdays. Conclusion We found no evidence of a higher risk of conversions, post-operative procedures, or readmission during weekends compared with weekdays. Despite this, a prolonged length of stay was observed in patients operated with cholecystectomy on Fridays and Saturdays. The observed difference could be due to ward rounds on weekends mainly focus on the sickest patients leaving less time for discharge.

Introduction A phenomenon ‘‘weekend effect’’ has previously been used to describe the higher risk of death and complications after surgery during the weekend compared to weekdays [1, 2]. The weekend effect is mostly seen in acute settings [3], but & Josephine Philip Rothman [email protected] Jacob Rosenberg [email protected] 1

Department of Surgery, Center for Perioperative Optimization, Herlev Hospital, University of Copenhagen, Herlev Ringvej 75, 2730 Copenhagen, Denmark

has also been proven for elective procedures [4, 5]. It is commonly described with diseases such as childbirth [6], diverticulitis [7], gastrointestinal bleeding [8], and in cardiology [9]. In addition, weekend time has also been described to lead to a longer length of stay (LOS) after operation for diverticulitis and hip fracture [10, 11]. Despite that elective laparoscopic cholecystectomy is often considered a safe outpatient procedure with a short convalescence and low risk of complications [12, 13], some patients are operated acute with a higher risk of complications. An increased risk of complications with nighttime cholecystectomy compared with daytime surgery, has previously been suggested [14]. The purpose of this study was to investigate if LOS differed throughout the

123

850

World J Surg (2016) 40:849–855

week and register if intra- and postoperative complications varied on weekend days compared to weekdays.

Methods This prospective database study was performed using a cohort from the Danish Cholecystectomy Database (DCD). The DCD was a nationwide mandatory registry existing from early 2006 to end 2011. It included patients from all surgical departments in Denmark, at both public and private hospitals, who for any reason had a cholecystectomy performed [15]. The DCD had a registration rate around 90 % from the start until mid 2011 [16]. Registrations in the DCD consisted of patient-related variables and operative findings entered into a secure website by the operating surgeon immediately following surgery. This was matched with register-based administrative data from the National Patient Registry including LOS, type of admission (emergency or elective), readmissions, and post-operative surgical procedures within 30 days. The DCD defined acute cholecystitis as clinical and ultrasonic pre-operative findings of acute cholecystitis combined with perioperative finding of edema in the gallbladder wall. Chronic cholecystitis required a thickened wall with fibrosis and dense adhesions judged by the operating surgeon. Previous pancreatitis was defined as an admission with the diagnosis 3 months prior to surgery. Previous upper abdominal surgery was defined as a scar in the area between the umbilicus and the xiphoid process. The cohort in this study included adults (C18 years) who had a cholecystectomy performed by either open surgery or standard four-port laparoscopy [17] in the registration period of the DCD. Patients treated with open surgery in accordance with Danish cholecystectomy guidelines [18] were excluded. Likewise, patients treated with single-incision cholecystectomy (SILS) or natural orifice transluminal endoscopic surgery (NOTES) technique were excluded. If data on operation type (open, laparoscopic, SILS, NOTES) or type of admission (emergency or elective) or other risk factors used for adjusting the outcome were missing, the patients were excluded. The outcomes chosen for this study were conversion rate, readmission rate within 30 days, post-operative surgical procedures within 30 days, and LOS. Two previously validated quality parameters, high-quality hospitalization process (defined as length of stay (LOS) B1 day and no readmission), low-quality hospitalization process (defined as LOS [3 days and/or readmission within 30 days), were used [19]. All outcomes were compared by weekdays (Monday through Friday) versus weekend days (Saturday and Sunday including public holidays on weekdays). The

123

Fig. 1 The association between day of surgery and length of stay (LOS) in days, illustrated with median, and error bar from 25th to 75th percentiles

LOS (median and quartiles) was stratified by the day of the week (Fig. 1). Statistics Outcomes were adjusted in a multivariate logistic regression analysis (using the enter method) for risk factors that were found significant in the so far largest published risk factor study on this subject (Table 1) [20]. Public holidays were grouped in the same category as Sundays, as the health care resources available in the Danish health care sector are comparable on these days. Results from the analysis were presented as odds ratios (OR) with 95 % confidence interval (CI). The multivariate logistic regression analysis was performed to determine if there was a higher risk of conversion, readmissions within 30 days, post-operative surgical procedures within 30 days, high-quality hospitalization process, or low-quality hospitalization process when being operated on the weekend days compared with ordinary weekdays. Median LOS and quartiles were used to describe LOS graphically stratified to each day of the week. Patients with a LOS of more than 1 day were grouped together. To determine if the day of the week was an independent risk factor for a prolonged LOS, a multivariate logistic regression was performed. All risk factors for a prolonged LOS (Table 1) as well as the specified weekday of operation were included in the multivariate model. This multivariate analysis was made for each of the days in the week (Table 2). A subgroup analysis for patients operated for acute cholecystitis was performed. p \ .05 was defined as significant.

World J Surg (2016) 40:849–855

851

Table 1 Known risk factors for each outcome Conversion

Length of stay

Post-operative procedures

Readmission

Sex

Yes

No

No

No

Age \60 years

Yes

Yes

Yes

No

ASA score \2

Yes

Yes

Yes

Yes

Acute cholecystitis

Yes

Yes

Yes

No

Chronic cholecystitis

Yes

Yes

Yes

No

Previous upper abdominal surgery

Yes

Yes

No

No

Previous pancreatitis

No

Yes

Yes

No

Open cholecystectomy

No

Yes

Yes

Yes

Surgical experience

Yes

No

No

No

Conversion rate was adjusted for acute cholecystitis, age, American Society of Anesthesiologists (ASA) risk score, chronic cholecystitis, previous upper abdominal surgery, sex, and surgical experience. Length of stay was adjusted for acute cholecystitis, age, ASA score, chronic cholecystitis, open cholecystectomy, previous upper abdominal surgery, and previous pancreatitis. Post-operative procedures within 30 days were adjusted for acute cholecystitis, age, ASA score, chronic cholecystitis, open cholecystectomy, and previous pancreatitis. Readmission within 30 days was adjusted for ASA score and open cholecystectomy Table 2 Risk of intra- and postoperative complications adjusted for known risk factors Monday to thursday

Friday adjusted OR (95 % CI)

Saturday adjusted OR (95 % CI)

Sunday/holiday adjusted OR (95 % CI)

Conversion

1

.91 (.8–1.04)

1.21 (.97–1.51)

1.09 (.86–1.38)

Readmission (30 days)

1

1.03 (.93–1.15)

1.07 (.85–1.35)

1.22 (.97–1.55)

Post-operative procedures (30 days) High-quality hospitalization process

1

1.02 (.88–1.18)

1.03 (.77–1.38)

1.12 (.83–1.50)

1

.85 (.79–.92)

.72 (.60–.87)

.88 (.72–1.07)

1

1.11 (.99–1.26)

.91 (.73–1.13)

1.03 (.82–1.29)

Low-quality hospitalization process

OR odds ratio, CI confidence interval Results in italics indicate statistical significance at (P \ 0.05)

Statistical analyses were performed using SPSS version 22 (IBM Corp., Armonk, NY, US). The study was approved by the Danish Data Protection Agency (Journal Number HEH-2013-078) and the DCD. No permission from the local ethics committee was needed according to Danish law.

Results A total of 28,759 patients were treated with cholecystectomy between January 2006 and December 2011 and met our inclusion criteria. A total of 8211 patients were excluded due to missing risk factors used for adjusted analyses. The majority of the included patients were completed laparoscopically, while some were converted to open surgery or primarily started as open surgery (against recommendations from the guidelines) (Fig. 2). The patients’ demographics and surgical findings are presented in Table 3. The greater part of cholecystectomies was

performed during the week, while only 4.7 % were performed during the weekend and public holidays. Patients operated on weekends and public holidays tended to be slightly older (54 years (40–68) vs. 50 years (37–62), p \ .05), had a higher ASA score (2 (2-2) vs. 1 (1-2), p \ .05), and a larger proportion of male patients were operated during the weekend compared to during the week (38.3 vs. 28.3 %, p \ .05). The underlying diagnoses that lead to surgery were more often acute cholecystitis during the weekend (70.8 vs. 16.7 %) while chronic cholecystitis and previous pancreatitis more often were operated during the week, respectively, 28 vs. 12.1 % and 6.6 vs. 2.8 % (Table 3). During the weekends, 17 % of the patients operated by laparoscopy were converted to open surgery compared to 7 % during the weekdays (OR 2.54 (2.18–2.95)) in the univariate analysis. However, in the adjusted analysis, the OR for conversion on weekdays versus weekend days was 1.17 (.76–1.08). No difference was observed in conversion when comparing Monday to Thursday with Friday or

123

852

World J Surg (2016) 40:849–855

Fig. 2 Study flow diagram for the trial

Saturday or Sunday (Table 2). Table 2 shows that the results were non-significant for readmission within 30 days, post-operative procedures within 30 days, and a low-quality hospitalization process as well. In general, there were higher rates of readmission and post-operative procedures in patients operated during the weekends, but when all known risk factors were included in the multivariate analysis, the differences disappeared. Although the patients did not have more complications during the weekends, they less often received a high quality hospitalization process on Fridays and Saturdays, but not on Sundays (Table 2). Patients with acute cholecystitis did not vary from the general tendency. Monday through Friday the median LOS was 1 day, but increased on Saturdays and Sundays to 2 days (Fig. 1). The multivariate logistic regression revealed that the risk of having a LOS of more than 1 day was higher on Fridays and Saturdays, but significantly shorter on Wednesdays and Thursdays compared to the rest of the week (Table 4). An interesting finding was that the risk of a longer LOS did not increase when operated on Sundays and public holidays. As the majority of patients operated on weekends were patients with acute cholecystitis, a subgroup analysis was performed on these patients. For this subgroup, the analysis revealed that the risk of having a LOS of more than 1 day was higher on Saturdays than any other day of the week (OR 1.47 95 % CI 1.15–1.86) (Table 4).

Discussion This study was a prospective database study on 28,759 patients who had a cholecystectomy performed in the period from 2006 to 2011. No differences in complications

123

were observed between whether the operation was performed on weekdays or weekends. Despite this, a longer length of stay was observed on Fridays and Saturdays compared to the rest of the week. Operation on Wednesdays and Thursdays resulted in a reduced length of stay. Patients with acute cholecystitis had a longer length of stay on Saturdays. Similar to the results of the present study, an analysis of multiple Danish database registers on elective surgeries (bariatric surgery, knee and shoulder and hip replacement surgery) did not find a weekend effect when comparing readmissions in hospitals performing elective surgery both during work days and weekend days [21]. It might therefore not be a question of qualifications, but more structural or organizational issues that determine length of stay after operation. An English study has on the contrary found a higher mortality risk during weekends compared with Mondays for elective surgery [4]. To explain the phenomenon ‘‘weekend effect’’, it has been suggested that the likelihood of undergoing a surgical procedure is lower at the end of the week and during the weekends while, delay to surgery would typically be longer [22]. In the current study LOS was defined as the postoperative LOS and a longer LOS prior to the surgery, can therefore not explain our findings. It has also been suggested that staff on duty during the weekends, including doctors and nurses, are less specialized. The hypothesis is that the discharge therefore will be postponed to during the week when more specialized staff are on duty. Another explanation for a longer LOS during weekends could be due to the fact that ward rounds in weekends in Denmark mostly are conducted on a need-todo basis. The result may be that the most sick patients are prioritized, which potentially could limit the opportunity

2322 (40.1) 1599 (27.6) 146 (2.5)

40–59

60–79

[80

381 (6.6) 403 (7.) 571 (9.9)

Previous pancreatitis

Conversion Readmission (30 days)

14 (.2)

Reoperation (deep bleeding) 633 (10.9)

30 (.5)

Laparoscopy

Low-quality cholecystectomy

23 (.4)

Percutaneous drainage of intraperitoneal abscess

3917 (67.7)

47 (.8)

Endoscopic stent placement

High-quality cholecystectomy

109 (1.9)

ERS

Frequent additional procedures

237 (4.1)

Previous upper abdominal surgery

3 (.1)

5 769 (13.3)

11 (.2)

4

1705 (29.5)

314 (5.4)

3

Chronic cholecystitis

2195 (37.9)

2

Acute cholecystitis

3262 (56.4)

1

ASA score

1660 (28.7)

58 (1.)

4175 (72.2)

20–39

Age \20

Female

Sex

Monday n (%)

607 (11.7)

3296 (63.4)

19 (.4)

18 (.3)

37 (.7)

44 (.8)

107 (2.1)

398 (7.7) 596 (11.5)

325 (6.2)

194 (3.7)

1472 (28.3)

958 (18.4)

6 (.1)

13 (.2)

325 (6.2)

2092 (40.2)

2765 (53.2)

164 (3.2)

1512 (29.1)

2029 (39.)

1459 (28.1)

37 (.7)

3664 (70.4)

Tuesday n (%)

683 (11.5)

3904 (66.)

21 (.4)

15 (.3)

32 (.5)

43 (.7)

110 (1.9)

468 (7.9) 599 (10.1)

403 (6.8)

269 (4.5)

1735 (29.3)

1021 (17.2)

3 (.1)

19 (.3)

316 (5.3)

2153 (36.4)

3428 (57.9)

173 (2.9)

1683 (28.4)

2353 (39.8)

1661 (28.1)

49 (.8)

4265 (72.1)

Wednesday n (%)

776 (12.5)

4207 (67.5)

19 (.3)

32 (.5)

27 (.4)

31 (.5)

113 (1.8)

451 (7.2) 667 (10.7)

404 (6.5)

284 (4.6)

1719 (27.6)

959 (15.4)

5 (.1)

11 (.2)

330 (5.3)

2388 (38.3)

3497 (56.1)

178 (2.9)

1785 (28.6)

2500 (40.1)

1724 (27.7)

44 (.7)

4469 (71.7)

Thurday n (%)

586 (13.7)

2624 (61.3)

11 (.3)

17 (.4)

11 (.3)

28 (.7)

96 (2.2)

315 (7.4) 464 (10.8)

309 (7.2)

169 (3.9)

1039 (24.3)

881 (20.6)

3 (.1)

21 (.5)

254 (5.9)

1676 (39.2)

2326 (54.3)

107 (2.5)

1238 (28.9)

1643 (38.4)

1247 (29.1)

45 (1.1)

3090 (72.2)

Friday n (%)

202 (28.3)

216 (30.3)

1 (.1)

1 (.1)

6 (.8)

12 (1.7)

21 (2.9)

121 (17.) 86 (12.1)

19 (2.7)

20 (2.8)

85 (11.9)

492 (69.)

2 (.3)

4 (.6)

56 (7.9)

300 (42.1)

351 (49.2)

49 (6.9)

243 (34.1)

251 (35.2)

165 (23.1)

5 (.7)

448 (62.8)

Saturday n (%)

Table 3 Patients’ demographics and outcome results split on day of the week with quantity of patients and percentage in parenthesis

154 (33.)

142 (30.5)

1 (.2)

2 (.4)

6 (1.3)

11 (2.4)

16 (3.4)

84 (18.) 68 (14.6)

15 (3.2)

20 (4.3)

55 (11.8)

351 (75.3)

2 (.4)

12 (2.6)

50 (10.7)

165 (35.4)

237 (50.9)

42 (9.)

154 (33.)

152 (32.6)

114 (24.5)

4 (.9)

281 (60.3)

Sunday n (%)

52 (31.7)

52 (31.7)

0 (.)

2 (1.2)

2 (1.2)

1 (.6)

1 (.6)

22 (13.4) 20 (12.2)

4 (2.4)

4 (2.4)

23 (14.)

108 (65.9)

0 (.)

3 (1.8)

17 (10.4)

61 (37.2)

83 (50.6)

17 (10.4)

54 (32.9)

63 (38.4)

30 (18.3)

0 (.)

100 (61.)

Public holiday n (%)

3693 (12.8)

18358 (63.8)

86 (.3)

117 (.4)

144 (.5)

217 (.8)

573 (2.)

2262 (7.9) 3071 (10.7)

1860 (6.5)

1197 (4.2)

7833 (27.2)

5539 (19.3)

24 (.1)

94 (.3)

1662 (5.8)

11030 (38.4)

15949 (55.5)

876 (3.)

8268 (28.7)

11313 (39.3)

8060 (28.)

242 (.8)

20492 (71.3)

Total n (%)

World J Surg (2016) 40:849–855 853

123

854

World J Surg (2016) 40:849–855

Table 4 Day of the week as risk factor for prolonged length of stay (LOS) for all patients and for patients with acute cholecystitis Weekday

OR for LOS [1 day (all patients)

OR for LOS [1 day (acute cholecystitis)

Monday

1.03 (.95;1.12)

.91 (.75;1.09)

Tuesday

1.03 (.95;1.12)

1.03 (.87;1.22)

Wednesday

.91 (.84;.98)

.92 (.78;1.09)

Thursday

.84 (.78;.91)

.88 (.74;1.04)

Friday

1.19 (1.09;1.30)

1.1 (.92;1.31)

Saturday

1.39 (1.15;1.68)

1.47 (1.15;1.86)

Sunday

.99 (.79;1.25)

.99 (.69;1.17)

Public holiday

1.38 (.94;2.01)

1.23 (.75;2.0)

OR odds ratio Results in italics indicate statistical significance at (P \ 0.05)

for discharge after an uneventful cholecystectomy. Denmark has during the last decade, with the research of Dr. Kehlet, experienced a paradigm shift in how a postoperative course should proceed [23]. This has resulted in a general tendency towards shorter admissions [24]. The health system in Denmark is part of the public sector, and there is therefore a bigger incentive to reduce costs in contradiction to privately funded health systems. As seen in Table 3, there were more complicated cases and more patients with acute cholecystitis operated during the weekends. This was adjusted for in the multivariate analysis, which strengthens the results, and might even give a more conservative estimate. The results were supported with results from the subgroup analysis on acute cholecystitis, which showed matching results. The limitation of this study was that the time of admission of the patient was unknown and it was therefore impossible to measure the interesting parameter ‘‘delay to surgery’’. Time of day of surgery is also unknown which means that we were unable to distinct between the diurnal variations. This study indicates that there is an opportunity to reduce the LOS on Fridays and Saturdays. The costs of admission today in Denmark is between 630 and 1270 US$ per day. Attention on the problem with delayed discharge might be enough to reduce the increased LOS during the weekends. In conclusion, there was in our database of laparoscopic cholecystectomies no evidence of weekend effect with regard to conversions, complications, or post-operative procedures. In spite of this, there was longer postoperative LOS when patients were operated on Fridays and Saturdays compared with weekdays. Compliance with ethical standards Conflict of Interest present study.

123

There are no conflicts of interest regarding the

References 1. Ananthakrishnan AN, McGinley EL (2013) Weekend hospitalisations and post-operative complications following urgent surgery for ulcerative colitis and Crohn’s disease. Aliment Pharmacol Ther 37:895–904 2. Bell CM, Redelmeier DA (2001) Mortality among patients admitted to hospitals on weekends as compared with weekdays. N Engl J Med 345:663–668 3. Aylin P, Yunus A, Bottle A et al (2010) Weekend mortality for emergency admissions. A large, multicentre study. Qual Saf Health Care 19:213–217 4. Aylin P, Alexandrescu R, Jen MH et al (2013) Day of week of procedure and 30 day mortality for elective surgery: retrospective analysis of hospital episode statistics. BMJ 346:f2424 5. Mohammed MA, Sidhu KS, Rudge G et al (2012) Weekend admission to hospital has a higher risk of death in the elective setting than in the emergency setting: a retrospective database study of national health service hospitals in England. BMC Health Serv Res 12:87 6. Bendavid E, Kaganova Y, Needleman J et al (2007) Complication rates on weekends and weekdays in US hospitals. Am J Med 120:422–428 7. McIsaac DI, Bryson GL, van Walraven C (2014) Elective, major noncardiac surgery on the weekend: a population-based cohort study of 30-day mortality. Med Care 52:557–564 8. Ananthakrishnan AN, McGinley EL, Saeian K (2009) Outcomes of weekend admissions for upper gastrointestinal hemorrhage: a nationwide analysis. Clin Gastroenterol Hepatol 7:296 9. Lairez O, Roncalli J, Carrie D et al (2009) Relationship between time of day, day of the week and in-hospital mortality in patients undergoing emergency percutaneous coronary intervention. Arch Cardiovasc Dis 102:811–820 10. Worni M, Schudel IM, Ostbye T et al (2012) Worse outcomes in patients undergoing urgent surgery for left-sided diverticulitis admitted on weekends vs weekdays: a population-based study of 31 832 patients. Arch Surg 147:649–655 11. Ricci WM, Brandt A, McAndrew C et al (2015) Factors affecting delay to surgery and length of stay for patients with hip fracture. J Orthop Trauma 29:e109–e114 12. Keulemans Y, Eshuis J, de Haes H et al (1998) Laparoscopic cholecystectomy: day-care versus clinical observation. Ann Surg 228:734–740 13. Bisgaard T, Klarskov B, Kehlet H et al (2002) Recovery after uncomplicated laparoscopic cholecystectomy. Surgery 132: 817–825 14. Phatak UR, Chan WM, Lew DF et al (2014) Is nighttime the right time? Risk of complications after laparoscopic cholecystectomy at night. J Am Coll Surg 219:718–724 15. Bardram L, Rosenberg J, Kristiansen VB (2005) The Danish Cholecystectomy Database–DCD. Ugeskr Laeger 167:2618–2620 16. Rothman JP, Burcharth J, Pommergaard HC et al (2015) The quality of cholecystectomy in Denmark has improved over 6-year period. Langenbecks Arch Surg 400:735–740 17. Malladi P, Soper N Laparoscopic cholecystectomy. In: UpToDate, Collins KA (Ed), UpToDate, Waltham, MA. (Accessed January 26, 2015) 18. Sundhedsstyrelsen. Referenceprogram for behandling af patienter med galdestenssygdom 2006 (updated 12/01/06). Available from: http://sundhedsstyrelsen.dk/publ/PUBL2006/plan/sfr/galdesten/ GALDESTENSSYGDOMME.pdf 19. Harboe KM, Anthonsen K, Bardram L (2009) Validation of data and indicators in the Danish Cholecystectomy Database. Int J Qual Health Care 21:160–168

World J Surg (2016) 40:849–855 20. Harboe KM, Bardram L (2011) The quality of cholecystectomy in Denmark: outcome and risk factors for 20,307 patients from the national database. Surg Endosc 25:1630–1641 21. ‘‘Weekendeffekt’’ pa˚ danske hospitaler: en analyse baseret pa˚ landsdækkende kliniske kvalitetsdatabaser. Available at: http:// www.rkkp.dk/ej-synlige/weekendeffekt-pa-danske-hospitaler/. [Accessed May 9, 2015] 22. Dorn SD, Shah ND, Berg BP, Naessens JM (2010) Effect of weekend hospital admission on gastrointestinal hemorrhage outcomes. Dig Dis Sci 55:1658–1666

855 23. Kehlet H, Wilmore DW (2002) Multimodal strategies to improve surgical outcome. Am J Surg 183:630–641 24. Ugeskriftet. Sygehusene udskriver ældre tidligere og tidligere. (1. September 2015). Available from: http://ugeskriftet.dk/nyhed/ sygehusene-udskriver-aeldre-tidligere-og-tidligere

123

Cholecystectomy During the Weekend Increases Patients' Length of Hospital Stay.

A higher risk of complications and mortality has previously been proven in selected settings. The purpose of this study was to investigate whether len...
565B Sizes 1 Downloads 14 Views