Female gender is independently associated with longer hospital stays following infra-inguinal bypass for peripheral arterial disease. A retrospective cohort study

Abstract

Aims

The association between gender and length of hospital stay following infra-inguinal bypass (IIB) surgery is unclear. While previous studies have reported gender disparities in length of hospital stay (LoS), the results are conflicting and could be attributable to other confounding factors. We undertook this cohort study to determine if there are any gender differences in length of hospital stay following infra-inguinal bypass for PAD after adjusting for well-known confounders.

Methods

A 3-year single-centre retrospective case notes analysis of all people undergoing IIB between 2017 and 2019. Rutherford stage, graft conduit, urgency of bypass, level of bypass, procedure details, baseline demographics, length of stay (LoS) and co-morbidities were collected and univariable associations with length of hospital stay were reported. Factors associated with increased LoS on univariable analysis were entered into a multivariable model.

Results

177 IIB were analysed with a median age of 70 [63–73] years, 124 (70.1%) were male and 89 (50.2%) had DM. A total of 78 (44.1%) were current smokers, and 100 (56.5%) underwent emergency procedures. The cohort included patients with Rutherford stage 3 (n = 41 (23.2%)), stage 4 (n = 48 (27.1%)), stage 5 (n = 86 (48.6%)) and stage 6 (n = 1 (0.6%)) disease. A total of 100 (56.5%) underwent emergency procedures. The conduits used were prosthetic (n = 62 (35%)), vein (n = 113 (63.8%)) and composite (n = 2 (1.1%)), and the level of distal anastomosis was above knee (n = 49 (27.7%)), below knee (n = 66 (37.3%)) and distal (n = 62 (35%). Baseline demographics did not differ by gender, and there were no differences in post-operative complications. The proportion of patients discharged to their usual place of residence without a package of care did not differ by gender (p = .387). However, length of stay for female patients was significantly longer than for male patients (9 [6–21] vs 7 [5–14] days, p = .021). Other factors associated with increased LoS on univariable analysis were emergency versus elective (p < .0001), Rutherford stage (p < .0001), bypass level (p = .001), bypass conduit (p = .001), post-operative complications (p < .0001) and discharge to rehab or home with package of care (p < .0001). Patients operated on by a female surgeon also had a longer hospital stay (14 [8–20] vs 7 [5–14], p = .011) than those operated on by a male surgeon. After multivariate adjustment for bypass urgency, level and conduit, Rutherford stage, presence of post-operative complications and discharge destination, female gender (RR 1.59 95% CI: 1.09–2.3, p = .017) was still associated with increased length of hospital stay.

Conclusions

Even after adjustment for well-known factors associated with length of hospital stay, female gender appears to be independently associated with significantly longer hospital stays. Further investigation into factors affecting gender differences could shed further light on this apparent difference.