Retrospective analysis of outcomes and prognostic factors in patients with subarachnoid hemorrhage from rupture of small intracranial aneurysms: A multicenter study of 421 cases

INTRODUCTION

Subarachnoid hemorrhage (SAH) is an uncommon clinical entity but carries a high social and healthcare impact due to its substantial morbidity and mortality rates. Despite major advances in diagnostic tools and neurosurgical treatment, these figures have not significantly improved over the past decades. It is estimated that only about 10% of patients with severe SAH survive without neurological or cognitive deficits, whereas approximately 10–15% die before reaching medical care.^[1] This condition predominantly affects individuals of working age (particularly between 30 and 60 years), representing a considerable socioeconomic burden and causing a devastating impact on family dynamics. Contemporary population-based studies and systematic reviews report a 30-day mortality after aneurysmal SAH ranging from 30% to 40%, despite advances in neurocritical care and aneurysm treatment.^[2] Given the severity of the clinical presentation and the potential for rapid deterioration, once the diagnosis is confirmed, it is essential to promptly establish the initial grade of severity using validated clinical scales such as the World Federation of Neurosurgical Societies (WFNS) scale and the Hunt and Hess (HyH) scale. These tools allow accurate risk stratification and guide therapeutic decision-making, as prognosis and clinical outcomes are strongly determined by the initial neurological status.^[3] Severe SAH is typically defined by a score of 4 or 5 on the HyH or WFNS scales, which correlates with morbidity and mortality rates approaching 50% during the first 30 days following hemorrhage.

The objective of the present study was to analyze 1-year clinical outcomes in patients with SAH secondary to rupture of aneurysms smaller than 5 mm and to identify the main prognostic factors associated with mortality or poor outcome in this Latin American cohort.

MATERIALS AND METHODS

A retrospective, multicenter regional study was conducted including public and private referral institutions from several Latin American countries. Participating centers were selected based on their ability to provide complete clinical, imaging, treatment, and follow-up data, with the aim of minimizing loss to follow-up in a retrospective setting, rather than to reflect regional treatment preferences. The study period extended from 2017 to 2023. All patients aged 18 years or older with a diagnosis of SAH secondary to rupture of an intracranial aneurysm measuring <5 mm in diameter were included. Data collection was performed using a standardized form specifically designed for this purpose, which was systematically completed by a single member of the research team in all cases to ensure consistency in data recording. Clinical and demographic variables were analyzed, including age, sex, arterial hypertension, smoking, diabetes mellitus (DM), substance use, and personal or family history of aneurysm and/ or SAH. Variables related to clinical presentation were also recorded, such as the HyH/WFNS scale, classifying SAH as mild (grades 1–2), moderate (grade 3), or severe (grades 4–5). Diagnostic error was defined as the failure to clinically recognize aneurysmal SAH at the initial medical consultation or during subsequent evaluations, resulting in inappropriate discharge or delayed diagnosis and treatment. This definition refers to a clinical misrecognition rather than a technical failure of neuroimaging interpretation. The modified Fisher computed tomography (CT) grading scale was also evaluated. Regarding aneurysm morphostructural characteristics, the following were documented: aneurysm type (saccular, dissecting, or fusiform); location (anterior or posterior circulation); size (in millimeters): transverse and longitudinal diameters, and maximum neck width. Based on these measurements, a subanalysis was performed categorizing aneurysms as <3 mm or between 3 and 5 mm. In addition, aneurysm morphology was classified based on angiographic findings as either a simple aneurysm (regular and smooth contour) or a complex aneurysm (irregular wall, presence of blebs, or multilobulated configuration). Treatment modality was categorized as surgical or endovascular, and treatment-related complications were recorded, along with those associated with hospitalization or the SAH itself. The cause and timing of death were analyzed at 1-year follow-up. Functional outcomes were assessed using the modified Rankin Scale (mRS) at 1 year. Patients with mRS ≤2 were defined as having a good outcome and those with mRS ≥3 as having a poor outcome. To minimize loss to follow-up, outcome assessment was limited to a 1-year follow-up period. Given the retrospective design and the extended inclusion period of 7 years, longer follow-up intervals would have substantially increased the risk of missing data. By restricting follow-up to 1 year, complete outcome data were available for all included patients, and no patients were lost to follow-up during this period. Qualitative variables were expressed as absolute frequencies and percentages, while quantitative variables were summarized using means or medians as appropriate. Comparisons between proportions were made using the Chi-square test or Fisher’s exact test. Mean comparisons between groups were performed using the Student’s t-test or non-parametric equivalents (Mann–Whitney U) for non-normal distributions. To identify independent predictors of mortality and poor clinical outcome at 1-year follow-up, a binary logistic regression model was applied. Before this, a univariate analysis was conducted, selecting as candidates those variables with a p < 0.25. The final multivariate model was built using a stepwise backward selection strategy. Statistical significance was set at p < 0.05. All collected information was handled in strict confidentiality according to current ethical standards for epidemiological research. Patient identity was managed exclusively by the treating physicians. Informed consent was obtained from all participants or their legal representatives when patients were unable to provide it themselves. The study was approved by the institutional ethics committees of all participating centers and by the corresponding regulatory bodies in each country.

RESULTS

A total of 421 patients were included during the study period. Five Latin American countries participated in the study, with the following distribution of cases: Colombia (42%), Argentina (38%), Uruguay (12%), Chile (6%), and Ecuador (2%). In total, eight specialized cerebrovascular reference centers were involved (five endovascular and three microsurgical centers).

The mean patient age was 55.4 years, with a predominance of females (60%), and 69% of patients were treated within the private healthcare system of their respective countries.

The main relevant comorbidities and baseline characteristics of the study population are summarized in Table 1.

Regarding the clinical presentation of the 421 patients with SAH, headache (61%) and vomiting (35%) were the most frequent symptoms at admission, followed by loss of consciousness (23%). Seizures, focal neurological deficits, cranial nerve palsy, and coma were less commonly observed.

Regarding the initial diagnostic classification, 58% of patients were categorized as having mild SAH according to the HyH scale and 60% according to the WFNS scale. A total of 10% of patients were classified as moderate SAH by HyH and 8% by WFNS. Finally, 32% of patients were categorized as having severe SAH according to both HyH and WFNS scales, as detailed in Table 2.

Regarding the imaging findings on CT, the modified Fisher scale distribution for the 421 patients with SAH is detailed in Table 3. A high Fisher grade (3–4) at the time of initial diagnosis was observed in 79% of patients.

Diagnostic error, defined as failure to recognize aneurysmal SAH at the initial medical consultation or during subsequent evaluations leading to inappropriate discharge, was observed in 11% of patients included in the cohort.

Concerning aneurysm topography, 90% were located in the anterior circulation and 10% in the posterior circulation. Among those in the anterior circulation, the three most frequent locations were the middle cerebral artery (23%), anterior communicating artery (22%), and posterior communicating artery (17%). A total of 92% were saccular aneurysms, and 43% exhibited irregular morphology.

In terms of size, all aneurysms measured ≤5 mm in diameter, with a mean size of 3.11 mm (range 0.5–4.9 mm). Sixty-five percent measured between 3 and 5 mm,while 35% were smaller than 3 mm.

Of the 421 patients who received invasive treatment for the aneurysm, 119 (28%) underwent open surgery, whereas the remaining 72% were treated with endovascular therapy. Among the 302 patients treated by embolization, 78% received coiling and 28% were treated with other endovascular devices.

Treatment-related complications, regardless of the technique used, occurred in 10% of cases, without statistically significant differences between treatment modalities (12% in the surgical group vs. 8% in the endovascular group; p = 0.194). When clinical outcomes were stratified according to treatment modality, no statistically significant differences were observed between surgically and endovascularly treated patients in terms of 1-year mortality (16% vs. 18%, p = 0.534) or poor functional outcome (mRS ≥3) at follow-up (30% vs. 31%, p = 0.861). After adjustment for baseline clinical severity and acute complications, treatment modality was not independently associated with mortality or poor clinical outcome at 1 year.

During hospitalization, 42% of patients experienced at least one medical complication, the most frequent being thrombotic and infectious events.

Sixty-five percent of patients developed at least one complication related to SAH. The most frequent was vasospasm (49%), followed by acute hydrocephalus (29%), cerebral edema (11%), intraparenchymal hematoma (8%), and rebleeding (3%).

Among the 124 patients with acute hydrocephalus, only 9% (11 patients) did not receive any specific treatment. Of those who did, initial therapeutic approaches included external ventricular drainage in 51% and continuous lumbar drainage in 49%. Only 8% of patients with acute hydrocephalus developed chronic hydrocephalus requiring definitive treatment, either through ventriculoperitoneal shunt or cisterno-jugular shunt placement.

The overall mortality rate in this cohort was 18%. All potential predictors of fatal outcomes within 1 year after SAH from ruptured aneurysms <5 mm were evaluated. The results of the univariate analysis are presented in Table 4.

In the univariate analysis, age was statistically significant, with a mean of 63 years among patients who died compared to 56 years among survivors. For the multivariate analysis, all variables that showed statistical significance at a level of p ≤ 0.25 in the univariate analysis were retained. Variables identified as independent predictors of mortality are presented in Table 5, along with their corresponding odds ratios (OR) and 95% confidence intervals (CIs 95%).

It is important to note that, in this multivariate analysis, rebleeding showed an OR of 19.546; however, this result should be interpreted with caution due to the wide CI, which reflects a low precision estimate.

A total of 130 patients (31%) experienced poor clinical outcomes, as defined in the Materials and Methods section, the development during follow-up of any of the following: Moderate to severe disability requiring assistance, due to the inability to walk or attend to basic bodily needs (mRS 3–6).

To identify patient characteristics that could predict or anticipate a poor clinical outcome at 1-year follow-up among patients with SAH due to rupture of aneurysms smaller than 5 mm, we analyzed the variables whose univariate results are presented in Table 6.

As a result of the multivariate analysis (identification of predictors of poor outcomes), based on the 20 variables that showed a p < 0.25 in the univariate analysis, the final model included 8 identified variables. Their corresponding OR and 95% CIs (CI 95%) are shown in Table 7.

As in the multivariate analysis for mortality at 1-year follow-up, it was confirmed that age over 65 years, severe SAH at presentation, aneurysmal rebleeding, acute hydrocephalus, and the occurrence of treatment-related or medical complications were independent predictors of poor outcome at 1 year.

In addition, two other factors were identified as being associated with a worse prognosis: A personal history of SAH and the presence of cerebral edema.

DISCUSSION

Although small intracranial aneurysms have traditionally been considered to carry a low rupture risk, this assumption is largely derived from cohorts of unruptured aneurysms and may not fully reflect real-world clinical practice. Several clinical series focusing on patients presenting with acute SAH have demonstrated that aneurysms smaller than 5 mm account for a substantial proportion of ruptured aneurysms. In a Korean cohort, nearly half of ruptured aneurysms measured <5 mm at the time of diagnosis, highlighting that aneurysm size alone is not a reliable predictor of rupture.^[4] These findings underscore the limitations of extrapolating rupture risk estimates from selected unruptured aneurysm populations, such as those included in ISUIA^[5], to patients presenting with acute hemorrhage. Importantly, most available data on small aneurysm rupture derive from high-income countries with well-established screening programs and early access to specialized neurovascular care. In contrast, Latin American populations remain largely underrepresented in this literature. Differences in sociodemographic characteristics, burden of vascular risk factors, access to timely neuroimaging, and delays in diagnosis or referral may contribute to distinct rupture patterns and clinical presentations in resource-variable settings. In addition to aneurysm size, other factors such as morphological complexity, presence of blebs, vascular risk factors, delayed diagnosis, and disparities in access to timely neurovascular care may further contribute to rupture risk, particularly in resource-limited settings. Our multicenter Latin American cohort provides real-world evidence supporting the concept that small aneurysms can and do rupture and that their clinical behavior and outcomes should be interpreted within the context of regional healthcare realities. In this context, disparities in access to specialized neurovascular care and timely treatment pathways may also influence clinical outcomes, including mortality and long-term functional status, beyond aneurysm-related factors alone.

Regarding the initial clinical condition, this study found that 32% of patients presented with severe SAH, a proportion comparable to international reports, where 20–30% of cases are classified as severe at presentation.^[6,7] This finding is clinically relevant since the initial neurological status remains one of the main prognostic determinants. In our analysis, patients with severe SAH had a fivefold higher risk of mortality and a threefold higher probability of poor functional outcome compared to those with mild disease. Mortality in the severe group reached 35%, compared with 10% among mild cases. These results are consistent with international data, which estimate a poor outcome rate of >40% in severe SAH, dropping below 20% in mild cases (grades I–II).^[8]

In our cohort, diagnostic error was observed in 11% of patients, a frequency comparable to that reported in the international literature over the past decade (approximately 5.5–14.7%).^[9,10] Although diagnostic error is a well-recognized clinical issue, it did not retain independent significance in univariate or multivariate analyses for mortality or poor functional outcome in our cohort.

At 1-year follow-up, overall mortality was 18%, slightly lower than prior reports ranging between 30% and 40%.^[2] This difference may be explained by the fact that our cohort consisted exclusively of Latin American referral centers for SAH management. Literature consistently shows that patients treated in high-volume, multidisciplinary neurovascular centers achieve better outcomes, especially when managed by experienced neurosurgeons or neurointerventionalists. Moreover, treatment in dedicated neurocritical care units or in hospitals handling >35 SAH cases per year has been associated with lower mortality compared with low-volume institutions (<10 cases annually).^[11]

In univariate analysis, treatment performed within the first 24 h was associated with lower mortality but worse functional outcome at 1 year. This apparent discrepancy likely reflects confounding by indication, as patients treated early often present with more severe clinical conditions, prompting urgent intervention, while baseline neurological injury ultimately drives long-term functional outcome. After adjustment for clinical severity and other relevant covariates, early treatment did not retain independent significance in the multivariate model, suggesting that its effect is largely mediated by initial disease severity rather than timing alone.

Independent predictors of mortality identified in our study included age >65 years, history of DM, severe SAH (WFNS 4–5), aneurysmal rebleeding, acute hydrocephalus, treatment-related complications, and medical complications during hospitalization. These findings align with previous studies, which highlight the HyH, WFNS, and Glasgow Coma Scale scores as reliable predictors of mortality,^[12] along with vasospasm, intraventricular hemorrhage, and intracerebral hematoma as determinants of prognosis.^[13]

Functional outcomes were assessed using the mRS at 1-year follow-up. Poor outcome was defined as mRS ≥3. In our cohort, 31% of patients experienced poor functional outcome, similar to international series. When stratified by initial severity, poor outcome occurred in 19% of mild cases and 55% of severe cases, consistent with international reports showing approximately 20% poor functional outcome in patients with mild SAH and up to 60% in those with severe presentation.^[7,14]

In the multivariate analysis, independent predictors of poor outcome included age >65 years, history of SAH, severe SAH (WFNS 4–5), aneurysmal rebleeding, acute hydrocephalus, cerebral edema, treatment-related complications, and medical complications during hospitalization. Although aneurysmal rebleeding emerged as a strong independent predictor of mortality and poor functional outcome, the wide CI observed in our multivariate model reflects the relatively low number of rebleeding events in the cohort. This limited event frequency reduces the precision of the estimated effect size and warrants cautious interpretation of the magnitude of the OR. Nevertheless, the direction and strength of the association are consistent with previous studies identifying rebleeding as one of the most powerful determinants of adverse outcome after aneurysmal SAH.^[11,13] Van Donkelaar et al. (2019)^[15] developed the SAFIRE score, a predictive model incorporating age, Fisher grade, WFNS score, and aneurysm size, providing an accurate and easily applicable tool. Almatter et al. (2018)^[16] identified age, initial clinical grade, rebleeding, parenchymal hemorrhage, and aneurysm location (particularly middle cerebral artery) as key prognostic factors. For mild SAH, Zijlmans et al. (2018)^[17] reported advanced age, delayed cerebral ischemia, pneumonia, and meningitis as predictors of poor outcome, whereas Mocco et al. (2006)^[7] found age, high pre-operative HyH grade, aneurysm size, and hyperglycemia as negative predictors in severe SAH. These findings are consistent with our results across the Latin American cohort.

In line with the prognostic factors identified in our multivariate models, although treatment modality has traditionally been considered a relevant determinant of outcome, in our cohort of ruptured aneurysms smaller than 5 mm, the choice between surgical clipping and endovascular treatment was not independently associated with mortality or poor functional outcome at 1 year. In contrast, the occurrence of treatment-related complications emerged as a strong independent prognostic factor, significantly increasing the risk of both mortality (p = 0.001, OR 4.09) and poor clinical outcome (p = 0.025, OR: 3.41). Together with baseline clinical severity and acute in-hospital complications, these findings indicate that prognosis in this specific subgroup is driven primarily by patient and disease-related factors and by the safety of the procedure, rather than by the treatment technique itself.

CONCLUSION

SAH due to rupture of aneurysms smaller than 5 mm remains a high-risk clinical condition, even when managed in specialized centers. In this Latin American cohort, 1-year mortality and functional outcomes were comparable to those reported in international series. Initial neurological severity at presentation, as assessed by the WFNS and HyH scales, emerged as a key determinant of both mortality and long-term functional outcome.

Independent predictors of mortality included advanced age, DM, severe SAH at admission, aneurysmal rebleeding, acute hydrocephalus, and the occurrence of medical or treatment-related complications during hospitalization. In contrast, poor functional outcome at 1-year follow-up was independently associated with advanced age, prior SAH, severe SAH, aneurysmal rebleeding, acute hydrocephalus, cerebral edema, and in-hospital medical or treatment-related complications.

These findings highlight the critical role of early comprehensive clinical assessment, prompt management of acute complications, and multidisciplinary care in high-complexity cerebrovascular centers. Furthermore, the observed rates of mortality and long-term disability underscore the need to strengthen secondary prevention strategies, improve early detection of complications, and develop management protocols tailored to the epidemiological and healthcare realities of Latin America.