Independent Survivors of Brain Bleed Still Struggle a Year Later, Swedish Study Finds

People who survive a burst brain aneurysm and are judged functionally independent a year later still report substantially more trouble with everyday activities than a matched general population, according to a nationwide Swedish study published in Acta Neurochirurgica. The study also found that a standard functional outcome measure, the Glasgow Outcome Scale Extended, misses many of these quality-of-life problems, particularly difficulties with usual activities and extreme anxiety or depression.

The research drew on a national prospective multicenter study that involved all neurosurgical departments treating aneurysms in Sweden, a country of roughly ten million inhabitants, between September 2014 and March 2018. Data from 634 patients with aneurysmal subarachnoid hemorrhage were analyzed one year after the bleed. Functional outcome was assessed with the Glasgow Outcome Scale Extended, or GOSE, dichotomized as unfavorable (grades 2 to 4) or favorable (grades 5 to 8). Health-related quality of life was measured with the EQ-5D-3L questionnaire, a generic instrument covering five domains: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression, each rated as no problems, moderate problems, or extreme problems, with the collapse of moderate and extreme termed “any problems.”

Of the 634 patients analyzed, 471, or 74.3 percent, had a favorable GOSE score, meaning they were independent in daily activities. Follow-up was conducted at a median of 445 days after the hemorrhage, with an interquartile range of 206 days. The favorable-outcome group had a median age of 56 years, was 72.6 percent female, and had been treated endovascularly in 67.7 percent of cases. To put the patients’ quality-of-life ratings in context, the researchers constructed sex- and age-matched reference groups from published data on 32,597 people living in Stockholm County who answered a survey in 2006, with a response rate of 61 percent. The matching was done by indirect standardization, adopting the same percentage distribution of sex and age groups as in the patient populations.

The headline finding concerned usual activities. Independent patients reported moderate problems with usual activities at 24.0 percent, compared with 10.9 percent in their matched general population, an odds ratio of 2.58 with a p-value of 10 to the power of minus 16 or smaller. Any problems in this domain were also far more common among patients, at 27.0 percent versus 12.5 percent, with the same odds ratio of 2.58. There was also a trend toward more extreme problems among patients, at 3.0 percent versus 1.6 percent, with an odds ratio of 1.83 and a p-value of 0.030, which did not reach the study’s stringent significance threshold.

That threshold was set deliberately high. Because 95 analyses were performed, the researchers set the significance level at alpha equal to 0.001, meaning only differences with p-values at or below 0.001 were counted as statistically significant. The difference in extreme anxiety or depression, reported by 5.3 percent of independent patients versus 2.8 percent of the matched population, an odds ratio of 1.94, had a p-value of 0.002 and was therefore described by the authors as a clear trend rather than a significant difference. Moderate anxiety or depression was reported equally in both groups, at 33.1 percent versus 29.8 percent.

Not every domain showed the patients at a disadvantage. Patients actually reported fewer mobility problems than their matched population, at 9.6 percent versus 14.7 percent, an odds ratio of 0.62 with a p-value of 0.002, again just short of the significance threshold. Extreme mobility problems were equally rare in both groups, at 0.2 percent each. For self-care, the odds of moderate problems were essentially identical, at 2.1 percent versus 2.0 percent, and extreme problems were also similar, at 0.6 percent versus 0.5 percent. The authors concluded that for these two domains, which most closely resemble the criteria evaluated by GOSE, the EQ-5D-3L adds no additional information beyond the functional measure for independent patients.

Pain and discomfort told a different story than earlier work had suggested. Although pain and discomfort was the most frequently reported domain overall, with 50.3 percent of independent patients reporting moderate problems, this did not differ significantly from the matched population’s 47.6 percent, and extreme problems were reported by 3.6 percent of patients versus 4.8 percent of the matched population. The authors stated that they found no statistically significant or clinically relevant difference in pain and discomfort between the independent group and the matched general population, a finding that contrasts with previous studies that highlighted pain as a prominent issue in this population. They speculated that patients with pain problems may be allocated to the unfavorable GOSE group, both because of the hemorrhage itself and because of unfavorable baseline characteristics such as higher rates of hypertension and older age.

Patients with unfavorable GOSE scores fared worse than their matched population in all levels and dimensions measured, with one exception: moderate problems with pain and discomfort were reported equally. Instead, 26.4 percent of the unfavorable-outcome patients reported extreme problems with pain and discomfort, compared with 5.3 percent of the matched general population, a difference the authors reported as significant at p below 0.001. Global perception of health-related quality of life, measured on the EQ visual analogue scale from 0 to 100, averaged 68.7 in the whole study group with a standard deviation of 19.4. Independent patients averaged 74.1, while the unfavorable-outcome group averaged 52.2.

The study has several limitations the authors themselves emphasized. Only patients admitted to a neurosurgical facility were included, so those who died early or were too unstable for transport were excluded. Of an estimated 1,800 to 3,000 eligible cases expected over the 3.5-year recruitment period, only 1,037 patients with given consent were included, and 153 of the 910 patients with a first rupture of a saccular aneurysm were lost to follow-up, with a further 14 not completing the questionnaire and 109 deceased at follow-up. The study did not adjust for hypertension or smoking, both of which were more common among patients than in the general Swedish population and both of which can negatively affect quality of life. As a cross-sectional study, it cannot describe changes over time. The reference population came from Stockholm County in 2006, a region with a younger average age, higher income, and higher educational level than the rest of Sweden, which the authors note may have led to underestimation of reported problems in the matched population, and the time gap between the reference data and the patient data may also skew comparisons.

The authors state that, to their knowledge, this is the first time EQ-5D-3L data from a national group of aneurysmal subarachnoid hemorrhage patients with favorable functional outcome have been compared with a sex- and age-matched general population. Their conclusion is that despite achieving favorable functional outcome, these patients frequently experience difficulties with usual activities as well as anxiety or depression, and that the EQ-5D-3L offers limited incremental insight beyond GOSE for mobility and self-care. They suggest that targeting health-related quality-of-life issues that functional outcome measures do not capture could help identify independent survivors who are not, in fact, doing as well as their GOSE grade implies.

Subject of Research: Health-related quality of life one year after aneurysmal subarachnoid hemorrhage in functionally independent survivors compared with a matched general population

Article Title: Targeting health-related quality of life issues not found with functional outcome measures a year after aneurysm subarachnoid hemorrhage in a Swedish national cohort

Article References: Aineskog, H., Baldvinsdóttir, B., Nilsson, O. G., Jakola, A. S., Kronvall, E., Engström, E. R., Klurfan, P., Alpkvist, P., Svensson, M., Hillman, J., & Lindvall, P. (2026). Targeting health-related quality of life issues not found with functional outcome measures a year after aneurysm subarachnoid hemorrhage in a Swedish national cohort. Acta Neurochirurgica, 168(1), Article 240. https://doi.org/10.1007/s00701-026-07059-z

Image Credits: AI Generated

DOI: 10.1007/s00701-026-07059-z

Keywords: aneurysmal subarachnoid hemorrhage, health-related quality of life, EQ-5D-3L, GOSE, functional outcome, Swedish national cohort, anxiety and depression, usual activities, survivors, neurosurgery, patient outcomes, matched population comparison