In Reply: Nighttime Treatment of Ruptured Intracranial Aneurysms are Associated With Poor Outcomes

To the Editor: We appreciate the thoughtful commentary by Huang et al regarding our recent article on the outcomes of ruptured intracranial aneurysm treatment based on the time of intervention. Dr Huang et al raised an important question about the patients diagnosed during nighttime hours whose treatment was deferred until daytime.1 In our study, we demonstrated that nighttime treatment was associated with poorer outcomes compared with daytime treatment.2 The classification into these groups was based on the time of treatment not on the time of arrival at the hospital. We agree that for patients with favorable clinical conditions, treatment may have been more frequently delayed to daytime compared with those with unfavorable clinical conditions. Supporting this argument, the mean delay from admission to treatment was longer for the daytime group of 14.3 hours (SD: 78.3) compared with the nighttime group mean delay of 6.4 hours (SD: 5.7), P = .04. Moreover, patients treated at night were more likely to present with intracerebral hematomas with midline shift (14.1% compared with 6.3% for the daytime group, P = .01). Although unilateral pupillary dilation at admission was not statistically different between the groups (7.7% for nighttime vs 2.7% for daytime, P = .58), the poorer outcomes associated with nighttime treatment remained significant in multivariate analysis, independent of factors reflecting disease severity, such as the World Federation of Neurological Surgeon scale and the presence of intracerebral hematomas. Conversely, a delay from admission to treatment was not associated with favorable outcomes. As suggested, we conducted an analysis comparing patients whose treatment was delayed until daytime with those treated on arrival. Because some patients were diagnosed and later referred to this tertiary center, we used admission time rather than the diagnosis time for this analysis. In our cohort, 30% of patients (123/410, 5 missing data) admitted during daytime hours received nighttime treatment whereas 77.6% of patients (59/76, 2 missing data points) admitted during nighttime hours were treated in the daytime. Table 1 presents the clinical and radiological characteristics, as well as the outcomes for 4 different categories: (1) patients admitted during the day and treated during the day (n = 287), (2) patients admitted during the day and treated during the night (n = 59), (3) patients admitted during the night and treated during the night (n = 17), and (4) patients admitted during the night and treated during the day (n = 123). These groups did not show statistically significant differences in major clinical characteristics, with multivariate logistic regression indicating that these 4 categories were not predictors of favorable outcomes (Table 2). TABLE 1. - Clinical, Radiological Characteristics and Outcomes in Function of Admission Time and Treatment Time Admitted during daytime, treated during daytime (n = 287) Admitted during daytime, treated during nighttime (n = 123) Admitted during nighttime, treated during daytime (n = 59) Admitted during nighttime, treated during nighttime (n = 17) Statistical analysisFisher exact test P = Clinical characteristics Age, mean (SD) 55.8 (13.0) 54.5 (13.1) 58.2 (11.3) 51.7 (18.4) WFNS grade IV or V 77 (26.8%) 35 (28.5%) 18 (30.5%) 7 (41.2%) 0.58 Unilateral pupilar dilatation during admission 7 (2.4%) 4 (3.3%) 4 (6.8%) 2 (11.8%) 0.15 Treatment endovascular 140 (48.8%) 62 (50.4%) 27 (45.8%) 10 (58.9%) Rebleeding before treatment 8 (2.8%) 4 (3.3%) 4 (6.8%) 0 (0.0%) 0.42 Delay admission to treatment, mean (SD) 16.1 (94.1) 10.1 (4.7) 5.7 (4.6) 6.9 (6.8) Radiological characteristics Fisher scale grade III or IV 212 (74.1%) 91 (74.0%) 44 (74.6%) 12 (70.6%) Aneurysm location Anterior circulation 200 (69.7%) 83 (67.5%) 43 (72.9%) 11 (64.7%) Posterior circulation 87 (30.3%) 40 (32.5%) 16 (27.1%) 6 (35.3%) Aneurysm size 6.9 (4.4) 6.0 (3.3) 7.2 (3.8) 8.6 (7.3) Intracerebral hematoma Without midline shift 59 26 9 3 With midline shift 16 (5.6%) 3 (2.4%) 8 (13.6%) 2 (11.8%) Unknown 2 0 0 0 Treatment characteristics Decompressive craniectomy 29 (10.1%) 11 (8.9%) 5 (8.5%) 1 (5.9%) 0.97 Outcomes Favorable (mRS 0-2) 204 (71.1%) 92 (74.8%) 35 (59.3%) 10 (58.8%) 0.12 Unfavorable (mRS 3-6) 83 (28.9%) 31 (25.2%) 24 (40.7%) 7 (41.2%) mRS, modified Rankin scale; WFNS, World Federation of Neurological Surgeon. TABLE 2. - Logistic Regression for Factors Associated With Favorable Outcomes (mRS 0-2) Factor Univariate analysis Multivariate analysisa P OR (95% CI) P OR (95% CI) Age >65 y <.001 0.35 (0.22-0.53) .14 0.49 (0.19-1.22) Gender .47 1.18 (0.77-1.81) Location anterior .28 1.26 (0.83-1.91) WFNS 1-3 <.001 5.52 (3.59-8.54) <.001 5.33 (2.96-9.73) Fisher scale 1-2 <.001 4.45 (2.52-8.39) .019 2.29 (1.17-4.71) Aneurysm size >7 mm <.001 0.58 (0.43-0.78) .032 0.68 (0.48-0.97) Endovascular treatment .88 1.03 (0.70-1.52) Off-hours treatment .18 0.76 (0.50-1.15) Nighttime treatmentb .023 0.55 (0.34-0.93) Preoperative hematoma <.001 0.30 (0.19-0.46) .018 0.53 (0.32-0.90) Decompressive craniectomy <.001 0.21 (0.11-0.39) .071 0.50 (0.24-1.05) Moment of admission and treatment Admitted: Day, treated: Day Ref Ref Ref Ref Admitted: Day, treated: Night .51 1.18 (0.73-1.93) >.99 1.00 (0.57-1.77) Admitted: Night, treated: Day .070 0.58 (0.33-1.05) .062 0.52 (0.16-1.84) Admitted Night, treated: Night .27 0.57 (0.21-1.63) .32 0.54 (0.12-1.09) Bold entires indicate statistically significant findings. mRS, modified Rankin Scale; OR, odds ratio; Ref, References; WFNS, World Federation of Neurological Surgeon.aAn interaction factor between age and WFNS and age and Fisher scale was included.bThe factor nighttime treatment was not included here because of the collinearity with the factor “moment of admission and treatment.” One challenge for understanding the reasons for treatment delays is that the definition of nighttime and daytime treatment was applied using strict mathematical cutoffs, but these may not fully reflect the clinical realities. We also note that the major concern with delaying treatment is the risk of rebleeding. However, in this study, this risk was low and was not significantly different between subgroups. In conclusion, this study does not advocate for systematically delaying treatment until daytime. Rather, it highlights the importance of evaluating each case individually to determine the appropriateness and safety of emergent treatment. When medically safe and appropriate, treatment during daytime may be preferable.