Corticosteroids for non-operative treatment of symptomatic Markwalder grades 1 and 2 chronic subdural hematomas: A retrospective study of 35 cases

INTRODUCTION

Subdural hematoma (SDH) is a collection of blood and blood degradation products in the subdural space.^[1-3] It primarily affects elderly patients and there is a history of recent trauma in around 2/3^rd of them, and most of them are trivial.^[1,2] The incidence of chronic SDH is increasing globally primarily due to two reasons: An increase in the aging population and the use of anticoagulants/antiplatelets.^[1,2] The common clinical features include cognitive impairment, gait disturbance, limb weakness, headache, or raised intracranial pressure symptoms.^[2,4] Standard treatment for chronic subdural hematoma (cSDH) consists of burr hole drainage (BHD). Non-operative treatment includes medications such as corticosteroids, tranexamic acid, and atorvastatin. Middle meningeal artery (MMA) embolization is a minimally invasive endovascular option for treating cSDH which has been described in the past two decades and is gaining popularity in recent years.^[1,2,4-9]

Studies suggest that inflammation is involved in the disease process and propagation of cSDH.^[3,4,10] Corticosteroids reduce cSDH growth by inhibiting inflammation, reducing vascular permeability, and preventing the abnormal accumulation of fluid in the subdural space.^[3,4,10-12] Few retrospective studies and one recent meta-analysis showed that adjuvant corticosteroids are effective in cSDH and also reduce recurrence rates after surgery.^[4,13-17] However, two recent randomized trials concluded that dexamethasone resulted in fewer favorable outcomes and more adverse events but fewer repeat operations than placebo.^[18-20] In this paper, we aim to report our experience in managing lower (grades 1 and 2) MGS score cSDH with corticosteroids as the primary non-operative modality.

MATERIALS AND METHODS

This was a retrospective 4-year study conducted in the Department of Neurosurgery, Kasturba Medical College, Manipal, India. As a protocol, all cases of cSDH with low GCS scores and significant neurological deficits (roughly corresponding to Markwalder grades 3 and 4), significant midline shift (MLS) (>1 cm), and/or complete basal cisternal effacement were operated. Only borderline cases (symptomatic, conscious, no/mild deficits [corresponding to Markwalder grades 1 or 2], ≤1 cm MLS with preserved cisterns) were prescribed steroids and were analyzed in this study. The decision to operate or prescribe steroids for such borderline cases was at the discretion of the attending surgeon. As many cases of SDHs may contain mixed components (with acute and chronic components), those cases with predominant acute hematomas (hyperdense on computed tomography [CT]) were not included for steroids but a small acute component (<25%) with predominant isodense or hypodense hematoma was not a contraindication for inclusion, as illustrated in Figure 1. All patients received dexamethasone as the steroid (either parenteral or oral) in tapering doses (4 mg thrice a day for 5–7 days, followed by 4 mg twice a day for 5–7 days and then 4 mg once a day for 5–7 days).

Close

Figure 1:

Computed tomography (CT) brain showing (a) right-sided mixed density subdural hematoma (predominantly hypodense with small acute component) with significant bilateral hemispheric edema and midline shift (thickness/midline ratio 0.6) in a 54-year-old male presented with headache. (b) Follow-up CT 4 weeks later showing >90% resolution.

Export to PPT

The following data were analyzed: Age, gender, symptomatology, history of trauma, mechanism of injury, co-morbidities, imaging findings (detailed later), duration of steroid treatment, outcomes, recurrences, and complications. On imaging, the following were noted: Laterality, thickness of hematoma, MLS, thickness/MLS ratio, basal cisterns, sulcal edema, and density of hematoma (hypodense, isodense, or mixed). On follow-up (FU) scans, we arbitrarily categorized the hematoma resolution (compared between pre and post-treatment thickness of hematoma) as follows: Fully resolved, good response (>75% reduction), and poor response (<75% reduction).

Statistical analysis was performed using the Statistical Package for the Social Sciences software (version 29.0). Qualitative variables were presented as frequency and percentages. Quantitative variables were presented as mean (range). Variables studied were age, gender, co-morbidities, recent trauma, laterality, hematoma density, and T/M ratio. Age and T/M ratio were dichotomized based on their mean values. As they were qualitative/categorical variables, the Chi-square test was used for statistical association. These variables were tested for their association with resolution on FU CT scans. A P < 0.05 was considered statistically significant.

RESULTS

After exclusions, a total of 35 cases (24 males and 11 females) were analyzed. The male-to-female ratio was 2.18:1. The mean age was 52.2 years (range, 21–78 years). Headache was the predominant symptom, seen in 31 cases. Others included altered sensorium, speech problems, mild limb weakness, and seizures. There was a history of recent trauma in 19 cases. Comorbidities were seen in 18 patients, and five patients were on anticoagulants/antiplatelet medications. One patient had a low platelet count because of idiopathic thrombocytopenic purpura. Based on MGS, there were 28 grade 1 and seven grade 2 cases.

CT scan was performed in 33 cases while in 2 cases, magnetic resonance imaging was done. There were 11 right-sided, 15 left-sided, and 9 bilateral hematomas. The hematomas involved the frontotemporal region in 15 cases and the frontotemporoparietal region in 20 cases. The mean thickness of the hematomas was 14 mm in unilateral hematomas and 11 mm in bilateral hematomas. The mean MLS was 6.1 mm. The mean T/M ratio was 2.2 (0.6–4.1). Basal cisterns were partially effaced in 22 cases and normal in 13 cases, while all cases had hemispheric cortical sulcal edema. The density of hematomas was as follows: Hypodense – 13, isodense – 9, and mixed – 13 cases. The mean duration of steroids was 14 days and was prescribed in tapering doses as described earlier.

Outcomes

Clinically, all had improvement in symptoms and the mean time to improvement was 4 days. Radiologically, FU scans were available for 32 cases. In three patients, a repeat scan was not performed as they were lost to FU for outpatient department consultation but were consulted on the phone regarding their clinical improvement. On FU CT, there were 20 fully resolved, seven >75% resolved, and five partially resolved hematomas. The mean duration of FU CT was 4.3 weeks (range, 4–6 weeks) and the median duration was 4 weeks.

On analyzing various parameters with outcomes, we noted that age, gender, co-morbidities, antiplatelets/anticoagulant medications, history of trauma, and basal cisterns did not have a significant association with statistical analysis. With regard to the density of hematomas, we found that hypodense hematomas had higher resolution rates (11 out of 12) than isodense (7 out of 9) and mixed hematomas (9 out of 11); however, it was not statistically significant. The only parameter we found statistically significant was the T/M ratio. The T/M ratio of <2.2 had significantly better resolution rates than the T/M ratio of >2.2 (P < 0.05). The mean FU was 7.2 months (range, 1–9 m). We had 1 recurrence, noted after 1 month of successful resolution (as noted on the FU scan). For the patient with recurrence, steroids were prescribed twice and he developed 2 episodes of urinary tract infection (UTI). Other than this, we noted no adverse effects of steroids. Of the 34 cases, 11 cases were managed on an outpatient basis and the mean hospital stay for the patients who were admitted was 2.5 days (range, 2–5 days).

Table 1 summarizes the clinical-radiological data of the study cohort. Table 2 summarizes the statistical analysis.

Table 1: Characteristics of the study cohort

Characteristic/variable	Number (n) (%)
Total number	35
M: F	24:11
Mean age (years)   Range (years)	52.2 21-78
Laterality   Right   Left   Bilateral	11 (26%) 15 (45%) 9 (29%)
Co-morbidities   Yes   No	18 (55%) 17 (45%)
Recent trauma   Yes   No	19 (55%) 15 (45%)
Antiplatelets/anticoagulants   Yes   No	5 (20%) 30 (80%)
Mean thickness (mm)	14 (range, 6-22)
Mean midline shift (mm)	6.1 (range, 3-11)
T/M ratio*   <2.2>   2.2	26/35 17 (60%) 9 (40%)
Basal cisterns   Normal   Partial effaced	13 (41%) 22 (59%)
Density of hematoma   Hypodense   Isodense   Mixed	13 (35%) 9 (26%) 13 (39%)
Radiological follow-up**   Fully resolved/>75% resolved   Partial resolution	27 5
Mean follow-up (months)	7.2

*Only for unilateral hematomas, **No repeat scan for 3 patients

Table 2: Summarizing the statistics.

Characteristic/variable	P-value
Age	0.68
Gender	0.24
Co-morbidities	0.81
Recent trauma	0.22
Antiplatelet/anticoagulant	0.65
Density	0.63
T/M ratio	0.01
Laterality	0.31
Basal cisterns	0.54

Chi-square test was used. P< 0.05 considered significant.

Figures 1-5 illustrate the representative cases of our study.

Close

Figure 2:

Computed tomography (CT) brain showing (a) left-sided chronic subdural hematoma (thickness/midline ratio 1.1) and hemispheric edema in a 58-year-old female with thrombocytopenia, presented with disabling headache and vomiting. (b) Follow-up CT 4 weeks later showing complete resolution.

Export to PPT

Close

Figure 3:

Computed tomography (CT) brain showing (a and b) chronic subdural hematoma (hypodense) with significant bilateral hemispheric edema and midline shift (thickness/midline ratio 1.2) in a 21-year-old male presenting with headache. (c) Follow-up CT 5 weeks later showing resolution of hematoma with normal cortical sulci.

Export to PPT

Close

Figure 4:

Computed tomography (CT) brain showing (a) bilateral chronic subdural hematoma (hypodense) with sulcal effacement in a 56-year-old male with slurred speech and headache. (b) Follow-up CT brain showing resolution with normal cortical sulci.

Export to PPT

Close

Figure 5:

Computed tomography (CT) brain showing (a and b) chronic subdural hematoma (mixed) with minimal edema, midline shift (thickness/midline ratio 2.8) in a 70-year-old male presenting with headache.

Export to PPT

DISCUSSION

BHD is undoubtedly the most common surgery performed by neurosurgeons globally and is one of the most gratifying surgeries too. Although a simple and straightforward procedure, various complications have been reported which include acute hemorrhage, seizures, non-surgical problems, and rarely mortality.^[2,21] Furthermore, the recurrence rate is around 10–40%, as published in various studies.^[2,7,22-24]

Steroids were the first drugs to be tried as medical management of chronic SDHs and have yielded varying success rates over the years.^{[4,5,14,17-19,25-29]} Other medical methods include tranexamic acid, atorvastatin, and angiotensin-converting enzyme inhibitors, of which tranexamic acid has recently gained attention, primarily as an adjuvant to surgical evacuation. Another non-surgical but minimally invasive modality that is gaining popularity is MMA embolization which is based on the theory of interrupting the blood flow to the cSDH membrane, thus promoting hematoma resolution.^[8-10,30-34]

Pathogenesis of cSDH is multifactorial. Proposed theories include microbleeds from the endothelial gap junctions, coagulation, profibrinolysis, inflammation, and angiogenesis. Recent literature has shown the critical role of inflammation in causing fluid and blood exudation from neovascularized subdural membranes. Concentrations of matrix metalloproteinase (MMP)-2, MMP-9, and vascular endothelial growth factor (VEGF) are significantly elevated in hematoma fluid, suggesting that the MMPs/VEGF system may be involved in the angiogenesis of cSDH. Other inflammatory and angiogenic molecules that have been found to be elevated include tissue plasminogen activator, bradykinin, tumor necrosis factor, placental growth factor, interleukins (IL-6 and IL-8), and platelet activation factor.^[10,35-39]

Steroids have anti-inflammatory effects by the following mechanisms: They affect the differentiation and function of macrophages and B- and T-cells; mediate transcription of inflammatory proteins, such as chemokines and cytokines; and modify capillary endothelial cells and tight junctions by regulating the expression of the occludin gene thus promoting impermeability of the blood-brain barrier and decreasing the permeability of the vessels in cSDH. Furthermore, the levels of MMP, a critical enzyme responsible for modulating inflammation, angiogenesis, and vascular permeability, have been found to be decreased with the usage of steroids.^[3,11,12,40] It seems therefore logical that steroids with their potent anti-inflammatory effects might benefit these patients. However, available data show mixed results for the use of steroids in cSDH.

Berghauser et al. analyzed five observational studies providing class III evidence that suggests that treatment with corticosteroids for cSDH might be beneficial in the treatment of cSDH.^[13] Thotakura and Marabathina prospectively studied 26 patients of SDH in which the average thickness of the hematoma was 21.9 mm and the mean MLS was 10.3 mm.^[15] For 3 days, dexamethasone was given and was continued if neurological condition improved or else, BHD was performed. They had successful outcomes in 11 patients and 15 required BHD. They concluded that female gender, lesser MLS, and less density (Hounsfield units) were noted to be associated with successful medical treatment.^[15] Holl et al. conducted a systematic search of studies up to January 2019, including Randomized Controlled Trials (RCTs) or observational studies that compared corticosteroids as monotherapy (C), corticosteroids as an adjunct to surgery (CS), and surgery alone (S).^[16] The outcome measures were good neurological outcome, need for reintervention, mortality, and complications. They found no differences in good neurological outcomes between treatment modalities. This meta-analysis suggested that the addition of corticosteroids to surgery might be effective in the treatment of cSDH. However, the authors stated that results are to be interpreted with caution in view of the risk of bias in the included studies.

Hutchinson et al. published the results of their multicenter, randomized, placebo-controlled trial (Dex-CSDH trial).^[18] The aim was to assess the effect of dexamethasone in patients with symptomatic chronic SDH. They compared a tapering 2-week course of dexamethasone with a matching placebo in symptomatic patients in addition to receiving standard care (which could include surgery). Patients with mild or severe symptoms were enrolled and none were asymptomatic. They concluded that treatment with dexamethasone resulted in fewer favorable outcomes and fewer repeat operations for re-accumulation but was associated with more adverse events than placebo. However, the two drawbacks of the study were that 94% of them underwent surgery (major limitation) and CT characteristics were not defined (thickness, type of hematoma, MLS, etc.).

After the Dex-CSDH trial, Shrestha et al. published the observations of their meta-analysis.^[20] They noted that steroids were associated with a lesser recurrence. However, there was no benefit of steroid treatment in cSDH compared with nonsteroid treatment in terms of mortality and treatment success. Furthermore, there were few but significantly increased risk of adverse events with steroid treatment.

The most recent trial was one published by Miah et al. in 2023.^[19] It was a multicenter, open-label, controlled, non-inferiority trial where symptomatic patients with chronic SDHs (Markwalder grades 1–3 and hypodense/isodense hematomas on CT) were randomly assigned to a 19-day tapering course of dexamethasone or burr-hole drainage. They noted lower functional outcomes in the dexamethasone group in terms of modified Rankin score, Markwalder grading, and Extended Glasgow Outcome Scale. They observed complications in 59% of patients in the dexamethasone group and 32% in the surgery group, and additional surgery was performed in 55% and 6%, respectively. The trial was terminated early and the collaborators concluded that dexamethasone treatment was not found to be non-inferior to surgery with respect to functional outcomes and was associated with higher complication rates.

Zhang et al. published the results of their study where they studied the effects of dexamethasone in the treatment of recurrent cSDH.^[17] They included 27 patients with symptomatic recurrent cSDH, of whom 24 were given steroids. Of them, 17 had successful treatment while 7 required re-surgery. The authors concluded that recurrent cSDH can be treated successfully and safely with dexamethasone. To study the role of steroids as an adjuvant treatment in chronic SDH, Zhao et al. conducted a systematic review and meta-analysis of RCTs.^[29] They included five eligible RCTs with a total of 1251 patients. The findings showed that the use of adjuvant glucocorticoid therapy can effectively reduce the recurrence risk of cSDH compared with placebo. However, the use of adjuvant glucocorticoids resulted in a significant increase in psychiatric symptoms but noted no significant differences in infection between the two groups.

In our study, we treated 35 symptomatic patients of chronic SDH with corticosteroids. Initially (for around 5 cases), we gave it for 3 weeks and then switched over to a 2-week duration and we did not find any significant difference in outcome rates. As noted in the results section, we had a clinical success rate of 100%, a radiological success rate of 82%, and only one recurrence. This is because we included only borderline cases which seemed equivocal in the type of management and all were symptomatic. In this paper, we have described a novel radiological parameter, the thickness/midline (T/M) shift ratio which can be a predictor of successful treatment. We observed that cases with a lesser T/M shift ratio had significantly better outcomes. This means that patients who had significant hemispheric edema in comparison to the thickness of hematoma [Figures 1-5] responded well as steroids are very potent anti-edema drugs. On the other hand, we saw that steroids may not be very beneficial in the case of thick hematomas with mild/minimal edema. One drawback of this parameter is that it can be used only in unilateral hematomas as there will be very mild MLS in cases of bilateral hematomas. We also feel that the radiological parameter (T/M ratio) needs to be validated in a larger cohort of patients.

The success rates in our study were different from the two recently conducted trials.^[18,19] In the first study by Hutchinson et al.,^[18] >90% of patients underwent surgery in addition to steroids, and CT parameters were not taken into consideration. In the second study (non-inferiority trial) by Miah et al.,^[19] imaging characteristics were not well defined and Markwalder grade 3 patients (severe focal neurological deficits or stuporous) were also randomized to either steroids or burr-hole drainage. We feel that considering the Markwalder grading, only grade 1 and 2 patients will be eligible for steroid treatment, and higher grades will require surgery in view of either greater thickness of hematoma compressing the tracts or significant edema and subsequent raised intracranial pressure. We did not include any higher grade (MGS 3,4) cSDH in our study and all such patients routinely underwent BHD at our center. Furthermore, we had a small sample size in our study primarily because of the fact that the decision to operate or prescribe steroids in borderline cases was at the discretion of the attending neurosurgeon. Without robust data regarding the efficacy of steroids in cSDH, many surgeons would opt for BHD. With regard to the adverse effects, except for one case of UTI, we did not find any adverse effects of steroids in our patients. One surprising finding was that there were two diabetic patients in the cohort (they did not consent to surgery) and they too did not have any major blood sugar variations. However, we do not advocate prescribing steroids in diabetics and if given, to be cautious and aggressively monitor blood sugar values for the first 5 days (as higher doses are given initially). In our experience, we feel that in selected symptomatic cases, steroids will definitely benefit and obviate the need for surgery without any major adverse effects. Few authors may advocate MMA embolization as a non-surgical method in treating such cases as it involves a single-stage procedure and no long-term medications. However, there are obvious drawbacks that include the non-availability of cath lab facilities in rural as well as few urban centers, radiation exposure, and the cost. Further, there are no level 1 recommendations for MMA embolization and it still remains one of the non-surgical (yet minimally invasive) options for treating chronic SDH. In our opinion, steroids are reasonable alternatives to either BHD or MMA embolization in patients with MGS scores 1 and 2.

In our study, none of the residual hematomas were symptomatic. Schucht et al. conducted a randomized trial to investigate the influence of routine CT scan after surgery for cSDH.^[41] The authors found no benefit in routine CT scans as compared to those performed only in cases of neurological deterioration or with persisting neurological deficits, and patients in the group with routine clinical FU had fewer surgeries. Nevertheless, we considered the residual hematomas on FU scans as a partial response in our study. Second, although we did not compare our cohort with other methods (including surgical drainage), we feel that medical treatment would definitely reduce the treatment costs as many of them were managed on an outpatient basis. Future trials should include both the clinical and radiological parameters into consideration and compare similar cases while considering randomization and the analysis should include the outcomes and the economics in both the modalities.

CONCLUSION

Surgery remains the gold standard treatment for chronic SDH. In selected (borderline) cases of lower Malkwalder grade cSDH, steroids can be given and good outcomes can be expected, and in our opinion, steroids are reasonable alternatives to either BHD or MMA embolization in such patients. We described a novel radiological parameter, the T/M ratio, which can be a good predictor of successful treatment. However, this needs to be validated in a larger cohort of patients. We noted clinical improvement in all patients and had only one recurrence and five asymptomatic partially resolved hematomas on FU CT scans. Well-designed multi-centric prospective trials including both clinical and radiological parameters and analyzing both the outcomes and economics need to be performed in the future.