Factors determining outcome of post-angiogram-negative subarachnoid hemorrhage.

Aim: To determine the risk factors affecting outcome at the end of 90 days of post-angiogram-negative subarachnoid hemorrhage (SAH).
Methods: Non-traumatic SAH cases were reviewed from the case records of patients who had reported to the Department of Neurology of a tertiary care hospital and 50 angio-negative SAH cases were included after excluding all the cases with known cause of hemorrhage after doing computed tomography angiography (CTA)/digital subtraction angiography (DSA). The presence of hypertension, diabetes mellitus, coronary artery disease (CAD), history of alcohol and smoking, and various scales like Hunt and Hess Scale (HHS), World Federation of Neurological Surgeons (WFNS), and Fisher scale had been recorded at admission. The outcome was assessed at 90 days post-SAH using the Modified Rankin Scale (mRS). Statistical analyses: The association between the outcome and the factors was assessed using the Pearson Chi-Square test and the risk factors/predictors of outcome were assessed using logistic regression.
Results: The following variables were important risk factors for predicting poor outcome of angio-negative SAH (mRS 3 to 6): hypertension (P = 0.011), diabetes mellitus (P = 0.032), being an alcoholic (P = 0.019), HHS grade 4 to 5 (P < 0.01), and WFNS grade 4 to 5 (P < 0.01). On multivariate regression analysis, hypertension (P = 0.032) was an independent predictor of unfavorable outcome. Conclusions: At time of admission, presence of hypertension, diabetes mellitus, history of alcohol consumption, and poor grades of HHS and WFNS scale are predictors of poor outcome of angio-negative SAH.

Introduction

Subarachnoid hemorrhage (SAH) is defined by the extravasation of blood into the subarachnoid space. The most common cause of SAH is trauma; amongst the non-traumatic cases, rupture of an intracranial aneurysm is the leading cause, representing up to 85% of cases.[1] Sometimes, no structural cause for the hemorrhage can be identified on radiographic imaging. These hemorrhages are termed idiopathic SAH or angiogram-negative SAH.[23] The incidence of angiogram-negative SAH has been reported to be varying from 2% to 24% in various studies.[4] The incidence of angio-negative SAH in India probably lies at the higher end of the range mentioned in the literature.[4]

There also appears to be a wide variation in the investigation, follow-up, and delayed detection rate of vascular abnormalities in non-aneurysmal SAH (naSAH) patients.[5] The gold-standard investigation is digital subtraction angiography (DSA), but it is an invasive technique with risks of complications.[6]

The risk factors for the outcome of SAH have been well established by many studies, but they do not differentiate specifically between aneurysmal SAH and angiogram-negative SAH.[7] Meta-analysis by Mohan M et al.[8] discussed that these patients with angiogram-negative SAH are often left with questions about their prognosis and the need for further testing and procedures. This is an era of involving the patient and family in the decision-making process and this presentation of the risk--benefit balance is an important aspect. Identification of high-risk patients who may have a poor outcome and need thorough investigation and management in neuroscience centers with complex neuro-rehabilitation protocols, as well as potentially low-risk patients who have a good outcome and can be treated with more conventional methods, is paramount need of the present times.[8] Resources limited settings warrant the use of strategies that are simple and which can be used at the basic healthcare level. The factors that are considered in the present study can be assessed by the available resources in terms of manpower and equipment below the tertiary health care level. A literature search revealed lacuna concerning the factors affecting the outcome of angiogram-negative SAH, especially in the Indian context. Therefore, the present study was done to determine the risk factors affecting outcome at the end of 90 days post-angiogram-negative subarachnoid hemorrhage (angio-negative SAH).

Methods

The present study was a cross-sectional longitudinal study done by carrying out a retrospective and prospective review of the case records of patients admitted with angio-negative SAH. Ethical clearance was obtained from Institutional Ethics Committee, Army Hospital Research and Referral (AHRR), New Delhi, India.

Patients aged above 18 years, with computed tomography (CT) or magnetic resonance imaging (MRI) confirmed and CTA or DSA negative SAH and those who were willing to give consent were all included in the study. For patients who were comatose, the consent was obtained from the authorized representative/next of kin (NOK) of the patient. Patients with a history of trauma related to the disease and those who have an aneurysm and confirmed by CTA or by DSA were excluded.

Records of all SAH patients from Jan 2016 to August 2018 and of prospectively enrolled patients, from Sept 2018 to Jan 2020 were systematically reviewed using a predesigned standardized proforma. The sample included all the subjects who met the criteria within the period.

Diagnosis of SAH was based on Non-Contrast Computerized Tomography (NCCT) head studies by neurosurgeon. CTA or DSA was done to confirm the angio-negative status and the same was confirmed by the neurosurgeon. Repeat DSA was done from 14 to 90 days. The standard institutionalized therapy at the tertiary care center was followed. The patients who remained angio-negative after repeat DSA were included in the final analyses. Data for the following factors were recorded: age, gender, alcohol, smoking, diet, hypertension, diabetes mellitus, CAD, GCS, HHS, WFNS, and Fisher scale. Subjects who were taking alcohol more than 14 units/week for males and 7 units/week for females were considered as alcoholics.

Outcome: MRS was recorded by the investigator at 3 months post-SAH in person. Patients were divided into two categories, good outcome (mRS 1 to 2) and poor outcome (mRS 3 to 6).

Statistical analyses

Statistical analyses were performed using Statistical Package for Social Sciences software (SPSS version 20.0). The Association between the outcome of angio-negative SAH and various factors were assessed by applying the Pearson Chi-square test. The strength of the association was assessed by applying Spearman's correlation test. Simple binary logistic regression was applied to factors having significant moderate to high correlation to determine the important risk factors affecting the outcome.

Any variable associated with poor outcome of angio-negative SAH with a P value ≤0.05 on univariate analysis was entered in a stepwise model using multivariate logistic regression analysis to find out the factor which is more strongly associated.

Results

Diagnostic workup results

A total of 294 non-traumatic SAH cases (36 were retrospective cases and 258 were prospectively enrolled), confirmed by NCCT, reporting to the Department of Neurology, AHRR were reviewed. A total of 50 patients with angio-negative SAH were finally included in the analyses. [Figure 1] MRS was recorded by investigator in person (n = 17), from medical reports (n = 20) or by telephone interview (n = 13).

Figure 1

Schematic representation of the diagnostic workup

Patient characteristics and clinical presentation results

The mean age of the population was 51 years. The majority of the study subjects were males (58%). The majority of the patients were having a GCS score of 15 (28 patients), only two patients were having a GCS score of three, and the rest between 4 and 14. The majority (80%) of the population had a good outcome (mRS 1 to 2) at the end of 90 days post-SAH. All the subjects having poor outcome were in the mRS 6 (death) category. None of the subjects had mRS score of 3, 4, and 5. There was no significant association (P > 0.05) between age, gender, systemic diseases like CAD, and the Fischer scale. [Table 1]. Hypertension, diabetes mellitus, alcohol, smoking, were all significantly associated (P < 0.05) with the outcome of angio-negative SAH. The association between the outcome and the scales: HHS and WFNS scales was highly significant (P < 0.001) [Table 1].

Table 1

Association between the various factors and outcome of angio-negative SAH

	Good (mRS 1-2) (n=40) n (%)	Poor (mRS 3-6) (n=10) n (%)	P value*	r**
Age			0.488	0.115
<40 yrs	6 (15%)	2 (20%)
41-50 yrs	16 (40%)	2 (20%)
51-60 yrs	10 (25%)	2 (20%)
61-70 yrs	8 (20%)	4 (40%)
Gender			0.160	-0.223
Male	21 (52.5%)	8 (80%)
Female	19 (47.5%)	2 (20%)
Systemic diseases
Hypertension	14 (35%)	9 (90%)	0.002#	-0.441##
Diabetes Mellitus	4 (10%)	4 (40%)	0.021#	-0.327#
CAD	4 (10%)	0 (0%)	0.571	0.147
Habits
Smoking	18 (45%)	8 (80%)	0.048#	-0.280#
Alcohol	14 (35%)	8 (80%)	0.010#	-0.363#
WFNS			< 0.01##	0.611##
Good grade (I-III)	38 (95%)	2 (20%)
Poor grade (IV-V)	2 (5%)	8 (80%)
HHS			< 0.01##	0.606##
Good grade (I-III)	36 (90%)	2 (20%)
Poor grade (IV-V)	4 (10%)	8 (80%)
Fisher Scale			0.207	0.281
Grade 1	4 (10%)	0 (0%)
Grade 2	9 (22.5%)	0 (0%)
Grade 3	7 (17.5%)	2 (20%)
Grade 4	20 (50%)	8 (80%)

*Pearson Chi square test, **Spearman correlation test, #P < 0.05 = significant, ##P < 0.01 = highly significant

The following variables were important risk factors for predicting poor outcome of angio-negative SAH: presence of hypertension (P = 0.011), presence of diabetes mellitus (P = 0.032), alcoholic (P = 0.019), HHS grade 4 to 5 (P < 0.01) and WFNS grade 4 to 5 (P < 0.01) [Table 2]. Hypertension, diabetes mellitus, and alcohol were included in the multivariate logistic regression model in a stepwise manner. The only variable “hypertension” (P = 0.032) remained significant as an independent predictor of unfavorable outcome [Table 3].

Table 2

Simple binomial logistic regression with outcome of angio-negative SAH as dependent variable and associated factors as independent variable

Variable	Odds ratio	95% CI	P value
Hypertension	16.714	1.917-145.76	0.011*
Diabetes Mellitus	6.000	1.172-30.725	0.032*
Alcohol	7.429	1.384-39. 866	0.019*
HHS grade (Grade 4-5)	36.000	5.592-231.776	< 0.01**
WFNS grade (Grade 4-5)	49.33	7.050-345. 214	< 0.01**

*P < 0.05 = significant, **P < 0.01 = highly significant

Table 3

Multivariate logistic regression of systemic factors affecting poor outcome of SAH (mRS 3-6)

Variable	Wald	Odds ratio	95% CI	P value
Hypertension	4.609	11.826	1.240-112.801	0.032*
Diabetes Mellitus	1.688	3.427	0.535-21.961	0.194
Alcohol	2.636	4.615	0.728-29. 232	0.104

*P<0.05 = significant

Discussion

In this present era of technological advancements even with available investigations, angiogram-negative SAH remains a clinical challenge in terms of mortality, the risk, and frequency of re-bleeding, and regaining a normal daily functional capacity.[910]

The results of the present study revealed that the presence of hypertension was the most important predictor of poor outcome at the end of 90 days post angio-negative SAH. The presence of diabetes mellitus, alcohol use, HHS grade 4 to 5, and WFNS grade 4 to 5 were also important risk factors. Similar results have been reported in the literature for non-aneurysmal SAH and contrary results were reported for aneurysmal SAH, wherein hypertension and smoking were not associated with the outcome.[10] Favorable outcomes were associated with younger age contrary to the results of the present study.[11] The present study is unique in the sense of considering the independent association of multiple factors with the outcome of only angio-negative SAH cases. No other study to the best of the author's knowledge especially in an Indian setup has considered all the above-mentioned factors together.

The age group between 40 and 49 years has been most frequently reported in the literature to be suffering from SAH of unknown origin,[12] which is similar to the results of the present study. Some other studies reported that elderly patients had a significantly higher risk for an unfavorable outcome.[1113] For aneurysmal SAH, an association between advanced age and poor outcome was observed.[13] The predominance of young age and male sex has also been reported.[1114]

In the present study, repeat DSA was done between 14 and 90 days post-SAH. Gupta et al.[15] performed a second DSA at 4 to 6 weeks after the first angiography, whereas Khan et al.[16] reported the mean interval of 23 days (range = 3–89 days) between initial and repeat angiography. Xu et al.[17] performed repeat DSA either 10–14 days after admission or 1 month after discharge. Hence from previous studies, it was observed that the time of repeat DSA varied, but it was rarely done beyond 90 days post-SAH and it seemed to differ depending on a variety of factors such as hospital policy, patient willingness, imaging modality availability, and the limitations of a retrospective assessment of records.[16] Most of the authors have recommended a repeat angiography,[1518] whereas very few have not recommended it.[16] The overall misdiagnosis rate for perimesencephalic SAH (PMN-SAH), defined as positive results on repeat DSA, was only 1.1%, which was likely lower than the risk of the DSA procedure itself, but the overall misdiagnose rate for nPMN-SAH was 12.5%.[17] Patients with non-perimesencephalic SAH (nPMN-SAH) who had negative initial findings were strongly advised to have a repeat DSA.[17] The misdiagnosis rate was 10.9% for the present study.

A similar period of 90 days to assess outcome was also followed in some other studies.[1019] Shephard RH et al.[19] reported that the re-bleeding and early mortality after SAH of unknown etiology was observed at 3 months, whereas in subsequent years the patients tend to have fewer re-bleeds, fewer disabilities, greater working capacity, and lower later mortality compared to that of patients with an aneurysm or Aterio Venous Malformations.[20] A relatively long period of follow-up in patients with angio-negative SAH has also been recommended.[12]

HHS grade 4 to 5 were predictors of poor outcome in the present study, which is contrary to results by Naidech AM et al.[20] who reported that two patients with a negative initial angiogram result experienced re-bleeding. Both of these patients had good neurologic grades (Hunt-Hess grades I and III) on admission, and both died in the hospital after re-rupture. The explanation for this perhaps could be that the aneurysm thrombosed after the initial SAH and was angiographically invisible and these cases lend support to the policy of performing additional angiograms after several days in suspicious cases.[20]

Contrary to the results of the present study were reported by Konczalla et al.[13] wherein patients with Fisher grade 3 bleeding pattern had a significantly higher risk for an unfavorable outcome and death. Perhaps the reason for the difference could be the different timings of assessment of outcome, which was at 6 months post- SAH in their study.[13] Better modified fisher grade was found to be a predictor of the excellent outcome of aneurysmal SAH.[21] Some other scales also predict the outcome based on the amount of blood.[22]

WFNS grade 4 to 5 was an important risk factor for the poor outcome of angio-negative SAH in this which is comparable to some other studies.[1323] WFNS grade 1 to 3 is also a predictor of excellent outcomes in aneurysmal SAH.[21] Similar to our results, presence of hypertension has been reported to be an indicator of poor prognosis in patients with normal angiograms and presenting with subarachnoid hemorrhage of unknown cause.[19] Patients with PMN SAH tend to be less hypertensive and their chances of having unfavorable outcome increase with the presence of hypertension.[12] The reason for poor outcome perhaps could be that patients with hypertension report concomitant diseases like coronary occlusion, heart failure, cerebral thrombosis, and primary cerebral hemorrhage.[20] Some other authors pointed out that the presence of hypertension may increase the probability of re-bleeding in patients with SAH of unknown origin,[12] whereas the presence of hypertension is a predictor of poor outcome of angio-negative SAH, the same was not true for aneurysmal SAH.[21]

Gupta SK et al.[15] found that the presence of diabetes and alcohol intake was also higher in patients with negative angiogram results as compared with aneurysm positive patients and perhaps these comorbidities may play an etiologic role in the causation of SAH in patients with spontaneous SAH and negative angiogram results. Diabetes mellitus and alcohol intake were found to be risk factors for poor outcomes of angio-negative SAH in the current study, but further studies are required to substantiate the same.

The present study showed that smoking was associated with the outcome of angio-negative SAH which is in line with the result of another study.[24] Smoking related microvascular changes may have an effect on the risk of non-aneurysmal SAH and the most common assumption is that PMSAH has a venous instead of an arterial origin.[25] As a result of vascular inflammation, oxidative stress, and increased wall shear stress caused by increased blood viscosity and volume, smoking causes many pathological effects in the endothelium. Since endothelium is found in both arteries and veins, smoking may damage them both.[2627]

The incidence of cardiac diseases is more in angiogram-negative SAH patients as compared with the aneurysm-positive SAH patients.[15] Korja et al. found that CAD was not a risk factor for non-aneurysmal.[24] However, further studies need to substantiate the results of the present study concerning the effect of CAD on the outcome of angio-negative SAH.

Patients who died before the second angiography due to rebleed had a 2-mm aneurysm of the anterior communicating artery, according to autopsy reports published in the literature. After three negative angiographies, one case was surgically investigated, and a calloso-marginal artery aneurysm was discovered.[18] Angio-negative SAH also could be because of an intracranial aneurysm that thromboses early after the bleeding and in these cases, at short term follow-up, the sack could frequently recanalize and re-bleed, whereas a late follow-up shows that re-bleeding is very rare.[18]

Due to a variety of conditions, including vasospasm, a mass effect caused by hematoma, thrombosis, or technical issues, initial DSA can yield false-negative results,[28] whereas repeat DSA and CTA are utilized less commonly but have slightly higher diagnostic yield than cervical spine and brain MRI which are more commonly used.[29] The overall outcome for non-aneurysmal spontaneous SAH is better than aneurysmal SAH, but nPM-SAH has a poorer eventual outcome compared to PM-SAH, therefore repeat DSA should be done for all cases of nPM-SAH, before labeling them as non-aneurysmal SAH.[30]

The data was collected from a single tertiary care center, which may limit the generalizability of the study. As some of the cases were retrospectively obtained from records the typical restrictions exist, such as the lack of data not documented initially in the medical records, which might have introduced information bias in the study. Also, the inherent limitation of mRS scale being subjective contributes to the limitation of the present study. The results of the study can further be strengthened with analytical designs like case-control studies, with bigger sample size and analyzing the outcome based on various other parameters like biochemical parameters, coagulation profile, viral markers, liver function tests, MRI, 2D-ECHO, and ECG which could not be considered in the present study. Also, antihypertensive drugs and antiplatelet medications which are possible confounders for the outcome may be considered in future research.

Conclusions

The presence of hypertension, diabetes mellitus, and habit of alcohol, HHS grade 4 to 5, and WFNS grade 4 to 5 are important risk factors for poor outcome (mRS 3 to 6) of angio-negative SAH. The presence of hypertension among the systemic factors emerged to be the most important predictor of poor outcome (mRS 3 to 6) of angio-negative SAH.

The scales like WFNS and HHS that predict the outcome of angio-negative SAH are simple and can be routinely applied even in hospitals where super specialties are not available. The factors that surfaced in the present study, are very commonly observed in the Indian population. Therefore, doctors need to be aware of this while dealing with patients of angio-negative SAH and refer the patients to higher centers without much delay, which will go a long way in preventing mortality and also minimize morbidity.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.