Advertisement

Weekend admission in upper GI bleeding: Does it have an impact on outcome?

      To the Editor:
      We read with great interest the article by Abougergi et al
      • Abougergi M.S.
      • Travis A.C.
      • Saltzman J.R.
      Impact of day of admission on mortality and other outcomes in upper GI hemorrhage: a nationwide analysis.
      on the impact of admission day on clinical outcome after upper GI bleeding. By using data from the 2009 National Inpatient Sample, they found that mortality and length of stay were similar in patients admitted on weekends and weekdays, although the former had lower endoscopy rates and higher hospitalization charges. Despite the undeniable merits of their study, the authors could not control for important confounders because of limitations inherent to the type of data available to them. Notably, severity of GI bleeding or the occurrence of bleeding after admission, which may be related to significant mortality, could not be adjusted for.
      We performed a retrospective study including all adult patients diagnosed with upper GI bleeding at our institution from 2009 to 2011 (N = 1192). Weekend admission was defined as 12.01 pm Friday to 11:59 pm Sunday because the in-house endoscopy unit effectively closes for new referrals at midday on Fridays. Rockall scores and the Charlson comorbidity index were calculated. Outcome variables included performance of early endoscopy (≤24 from admission), endoscopic therapy, and in-hospital mortality (Table 1).
      Table 1Clinical data in patients presenting with upper GI bleeding on weekends versus weekdays
      All patients (N = 1192)Patients undergoing endoscopy (n = 916)
      Weekday (n = 770)Weekend (n = 442)P valueWeekday (n = 606)Weekend (n = 310)P value
      New admission, % (n)83 (638)87 (366).06485 (516)92 (284).004
      Age, y, mean (SD)72.1 (16.2)72.1 (17.5).9569.6 (15.8)69.8 (16.9).89
      Male sex55 (426)59 (247).2959 (355)61 (189).49
      Charlson comorbidity index (SD)1.90 (1.90)1.76 (1.82).231.74 (1.79)1.59 (1.71).2
      Presentation
       Shock, % (n)41 (316)36 (174).08441 (241)44 (132).36
       Hemoglobin, g/L, mean (SD)102 (28)106 (30).017101 (28)103 (28).15
       Rockall score, mean (SD)
      Before endoscopy.
      2.91 (1.67)2.85 (1.68).552.72 (1.64)2.63 (1.63).4
      Medications
       Aspirin, % (n)34 (255)36 (148).5534 (197)38 (115).23
       NSAID, % (n)10 (77)11 (46).6611 (66)12 (37).7
       Warfarin, % (n)12 (90)10 (43).4112 (73)9.6 (30).25
       Proton pump inhibitors, % (n)27 (198)23 (94).1626 (152)20 (61).049
      Endoscopic diagnosis
       Variceal bleeding, % (n)8.9 (54)7.7 (24).55
       Rockall score, mean (SD)
      After endoscopy.
      4.03 (2.05)3.74 (2.09).36
      Outcome variables
       Endoscopy, % (n)79 (606)74 (310).04
       Endoscopy within 24 h, % (n)53 (406)29 (121)<.00167 (406)39 (121)<.001
       Therapeutic endoscopy, % (n)26 (198)18 (75).00233 (198)24 (75).008
       RBC transfusion, % (n)66 (501)61 (255).168 (406)68 (209).9
       Surgery, % (n)2.5 (19)1.7 (7).362.8 (17)2.3 (7).62
       Angiography, % (n)2.6 (20)1.4 (6).183 (18)1.9 (6).35
       Surgery/angiography/TIPSS, % (n)5.1 (39)2.6 (11).0435.9 (36)3.5 (11).12
       Rebleeding, % (n)5.7 (44)4.3 (18).287.3 (44)5.8 (18).41
       Length of stay, d, mean (SD)8.3 (12.9)8.1 (12.6).778.3 (11.1)8.7 (14.1).67
       In-hospital mortality, % (n)11 (88)11 (48).987.1 (43)5.2 (16).26
      NSAIDs, Nonsteroidal anti-inflammatory drugs; RBC, red blood cell; TIPSS, transjugular intrahepatic portosystemic stent shunt.
      Data are reported as mean (SD) or % (no.) as appropriate. Significant differences (P < .05) are indicated in bold.
      Before endoscopy.
      After endoscopy.
      Although there were no significant differences in baseline data between patients presenting on weekends or weekdays, the former were less likely to receive early endoscopy or endoscopic therapy (Table 1). The 2 groups did not differ significantly in in-hospital mortality or length of stay. These results persisted after adjustment for confounders (age, sex, Rockall score, shock at presentation, the Charlson comorbidity index, new admission, variceal bleeding, and use of medications shown in Table 1) by means of logistic regression analyses, in which all potential confounders were forced into the final models (P > .05). After adjustment for confounders, weekend admission was still independently related to delayed endoscopy (>24 hours from presentation; P < .05), but not to the performance of endoscopic therapy (P > .05).
      In summary, the findings of Abougergi et al,
      • Abougergi M.S.
      • Travis A.C.
      • Saltzman J.R.
      Impact of day of admission on mortality and other outcomes in upper GI hemorrhage: a nationwide analysis.
      along with those of previous studies,
      • Nahon S.
      • Pariente A.
      • Latrive J.P.
      • et al.
      Weekend admission does not influence the mortality of upper gastrointestinal bleeding caused by peptic ulcers: results of a French prospective study of the association nationale des gastroenterologues des hopitaux generaux group.
      • Myers R.P.
      • Kaplan G.G.
      • Shaheen A.M.
      The effect of weekend versus weekday admission on outcomes of esophageal variceal hemorrhage.
      • Jairath V.
      • Kahan B.C.
      • Logan R.F.
      • et al.
      Mortality from acute upper gastrointestinal bleeding in the United Kingdom: does it display a “weekend effect”?.
      • Soncini M.
      • Chilovi F.
      • Triossi O.
      • et al.
      Weekend effect in non-variceal upper gastrointestinal bleeding: data from nine italian gastrointestinal units.
      suggest that there is no major weekend effect on patient outcome despite an observed delay to endoscopy at weekends. We have been able to confirm this in our data, which allowed adjustment for confounders not available in the study of Abougergi et al.
      • Abougergi M.S.
      • Travis A.C.
      • Saltzman J.R.
      Impact of day of admission on mortality and other outcomes in upper GI hemorrhage: a nationwide analysis.

      References

        • Abougergi M.S.
        • Travis A.C.
        • Saltzman J.R.
        Impact of day of admission on mortality and other outcomes in upper GI hemorrhage: a nationwide analysis.
        Gastrointest Endosc. 2014; 80: 228-235
        • Nahon S.
        • Pariente A.
        • Latrive J.P.
        • et al.
        Weekend admission does not influence the mortality of upper gastrointestinal bleeding caused by peptic ulcers: results of a French prospective study of the association nationale des gastroenterologues des hopitaux generaux group.
        Clin Gastroenterol Hepatol. 2009; 7 (author reply 912): 911
        • Myers R.P.
        • Kaplan G.G.
        • Shaheen A.M.
        The effect of weekend versus weekday admission on outcomes of esophageal variceal hemorrhage.
        Can J Gastroenterol. 2009; 23: 495-501
        • Jairath V.
        • Kahan B.C.
        • Logan R.F.
        • et al.
        Mortality from acute upper gastrointestinal bleeding in the United Kingdom: does it display a “weekend effect”?.
        Am J Gastroenterol. 2011; 106: 1621-1628
        • Soncini M.
        • Chilovi F.
        • Triossi O.
        • et al.
        Weekend effect in non-variceal upper gastrointestinal bleeding: data from nine italian gastrointestinal units.
        Am J Gastroenterol. 2012; 107: 635-636

      Linked Article

      • Impact of day of admission on mortality and other outcomes in upper GI hemorrhage: a nationwide analysis
        Gastrointestinal EndoscopyVol. 80Issue 2
        • Preview
          Studies have reached varying conclusions regarding the association between day of admission and outcomes in patients with upper GI hemorrhage (UGIH).
        • Full-Text
        • PDF
      • Response:
        Gastrointestinal EndoscopyVol. 81Issue 5
        • Preview
          We appreciate the interest expressed by Kalaitzakis et al in our study, as well as their observations regarding the impact of weekend admission on outcomes in patients with acute upper GI bleeding admitted to their institution. As with our study, they found no difference in inpatient mortality rates, endoscopic therapy rates, or length of stay between patients admitted on weekends and those admitted on weekdays. In addition, they confirmed the finding that patients admitted on weekends were less likely to receive an in-hospital endoscopy or early endoscopy (within 24 hours of admission).
        • Full-Text
        • PDF