Patient characteristics and outcomes of left ventricular assist device implantation on early versus late weekdays

, deﬁnitions. 2 Patient characteristics, described using proportions or median and quartiles, were compared between early to late weekday cohorts using Wilcoxon rank sum, c 2 , or Fisher exact tests. Time to mortality or debilitating stroke was censored if the LVAD was explanted for recovery, exchange or transplant, and compared up to 1-year postimplant using Kaplan–Meier analysis with log-rank tests. The hazard ratio (


RESULTS
Of 117 patients included, 66 and 51 patients were implanted on early and late weekdays, respectively.Patient characteristics were comparable (Table 1).Four additional patients implanted on Saturday were excluded.
Table 1 shows outcomes and complications during the index hospitalization.Patients in the late week cohort had longer post-LVAD LOS (median 18 vs 14 days, P ¼ .04).Other complications analyzed including in-hospital mortality, major bleeding, major infection, right heart failure, renal dysfunction, and stroke did not differ significantly.
At 1-year postimplantation, more patients in the late weekday cohort experienced renal dysfunction (20% vs 2%, P ¼ .001),whereas other complications did not differ significantly.(Table 1).Eighteen patients died and/or experienced a debilitating stroke, of whom 6 (9%) and 12 (24%) were in the early and late weekday cohorts, respectively.Survival free from debilitating stroke was lower in the late weekday relative to early weekday cohorts (P ¼ .029,There is lower event-free survival for patients implanted with an LVAD late in the week.

CENTRAL MESSAGE
Patients implanted with an LVAD late in the week may be more likely to have renal adverse events, longer lengths of stay, and lower event-free survival than those implanted earlier in the week.
stratified by device type P ¼ .037; Figure 1) with a 2.85 increase in the risk of the composite outcome (HR; 95% confidence interval [CI], 1.07-7.59).The majority (11/18) of the composite outcome events occurred within the first 3 months post-VAD.Complications occurring within the first 3 months that were associated with the composite  weekdays had (1) longer LOS post-LVAD, (2) more renal adverse events at 1 year, and (3) lower rates of survival or debilitating stroke than those implanted earlier in the week.
For those implanted later in the week, postoperative care within the first 48 hours of implant may be impacted by a "weekend" effect.Optimal level of care may be harder to achieve on the weekend due to lack of staffing and resource limitations during off hours with "on-call" and "more-experienced" staff available only for emergency situations. 3hus, our late weekday patients' experience and outcomes may show similarities with those admitted over the weekend.
A weekend effect on renal outcomes was also reported previously in patients with CS. 4 Renal dysfunction occurred early on at 3 months and persisted at 1 year following implantation.Worsening peri-and postoperative renal function is perhaps a surrogate of sicker patients not detected by our study and warrants further exploration.
Currently, one of the leading methods to improve outcomes associated with CS is improvement of treatment protocols. 5At our center (Baylor University Medical Center, Dallas, Tex), we have recently added 24/7 in-house heart failure attending coverage (in addition to 24/7 in house critical care attending coverage), daily multidisciplinary team rounding, and having system-wide protocols to guide timely transfers into the cardiac intensive care unit when warranted.In the setting of LVAD placement, the availability of expert care may eliminate deficits in patient care, allow for definitive planning regardless of implantation time and further provide immediate intervention for acute complications.Although we realize that dual in-house 24/ 7 attending coverage is not possible in all institutions, we propose that each institution periodically review their outcomes and modify protocols based on resources available to deliver optimal patient care regardless of the day of the week. 5here are several limitations to the study, including a retrospective single-center investigation with a modest sample size, so our results may not be generalizable to other centers.The sample size was sufficient to detect a difference in the primary outcome but may have limited power to detect smaller rate differences (eg, complication rate differences smaller than 25% when one group has a rate of 50%) between groups.In addition, some factors including team experience, on-call schedule, and staff fatigue may influence patient outcomes but are not captured in our medical record and thus were not included in this analysis.

CONCLUSIONS
Patients implanted with LVADs late in the week may be more likely to have renal adverse events at 1 year and longer length of stay following LVAD implantation with lower event-free survival than those who received an LVAD earlier in the week.Analysis of the full Intermacs database to ascertain if these results are generalizable is warranted.

FIGURE 1 .
FIGURE 1.Comparison the Kaplan-Meier curves with 95% confidence bands of the primary composite outcome of event-free survival to recovery or transplant up to 1-year postimplant between early-versus late-week LVAD implants (A) overall and (B) stratified by device type.The primary outcome is a composite of mortality or debilitating stroke (modified Rankin score >3).LVAD, Left ventricular assist device.