Comparisons of clinical characteristics, brain MRI findings, and responses to epidural blood patch between spontaneous intracranial hypotension and post-dural puncture headache: retrospective study

Spontaneous intracranial hypotension (SIH) and post-dural puncture headache (PDPH) are both caused by a loss of cerebrospinal fluid (CSF) but have different pathogeneses. Orthostatic headache is a cardinal symptom that worsens when the patient is upright and improves upon lying down. SIH is caused by CSF leakage from the spinal dura mater without a head trauma history or dural puncture [1]. In contrast, PDPH is caused by injuries or complications of medical procedures such as a lumbar puncture or epidural injection [2].

Brain magnetic resonance imaging (MRI) in patients with SIH can show any combination of pachymeningeal enhancement [3], sinus venous distention [4], subdural fluid collection [5], and pituitary hyperemia [6]. Brain MRI has been used for the diagnosis of SIH because of its ability to identify characteristic abnormalities. However, few reports have described the brain MRI findings for PDPH because this condition is mainly diagnosed based on procedure history and clinical features [7]. Moreover, the imaging findings of SIH and PDPH have rarely been compared.

Although many SIH cases show spontaneous resolution with conservative treatment, autologous epidural blood patch (EBP) injections may be required occasionally. EBP was initially introduced for PDPH treatment but is now also considered as the treatment of choice for SIH. However, the success rate of the first EBP for SIH is only about 30% ~ 60%, and multiple EBPs are frequently required [1, 8, 9]. On the other hand, most cases of PDPH also improve with conservative treatment, and EBP is usually recommended if the symptoms persist for more than ten days [10]. However, in practice, EBP is sometimes performed sooner to facilitate faster resolution of symptoms [11]. Thus, EBP is the currently preferred treatment for PDPH, and the symptoms resolve in about 75% of the cases [12].

SIH and PDPH are both caused by intracranial hypotension, but their pathogenesis and treatment policies differ depending on the disease. To help physicians plan the appropriate treatment approach (conservative treatment or EBP) depending on the disease, we investigated the differences in clinical characteristics, neuroimaging data, and responses to EBP between SIH and PDPH patients.

A total of 105 patients (45 men and 60 women; median age, 39 years [IQR, 32–52 years]) who met the inclusion criteria were reviewed. The median headache intensity (numeric rating scale) was 7 (IQR, 5.75–8), and the median EBP count was 1.0 (IQR, 0–2). Sixty-one patients (58.1%) had SIH while 44 patients (41.9%) had PDPH.

This study’s primary finding was that patients with SIH more often received EBPs and more often needed multiple EBPs than patients with PDPH. Most brain MRI findings were similar between SIH and PDPH patients, but the vG/SS angle was significantly different between the two groups.

The SIH and PDPH patients showed no significant differences in clinical characteristics, including hospitalization period. Nausea and vomiting were the most frequent symptoms in patients with SIH and PDPH. The interval between symptom onset and diagnosis was longer in SIH patients, which could be attributed to the fact that PDPH is diagnosed more quickly since the symptoms occur abruptly after the procedure. Although not statistically significant, auditory symptoms were more frequent in SIH patients than in PDPH patients. Because cochleovestibular manifestations occur due to traction or compression of the eighth cranial nerve or a reduction in the pressure of the perilymph, auditory symptoms were observed more often in patients with SIH, which represents chronically advanced intracranial hypotension [1]. Although patients with PDPH received significantly fewer EBPs than those with SIH, they showed no difference in hospitalization period and headache intensity. Our PDPH patients may represent a more severe subgroup of PDPH since we only included admitted PDPH patients, resulting in the lack of a difference in clinical severity between PDPH and SIH.

The characteristic MRI findings for intracranial hypotension include diffuse pachymeningeal enhancement, venous engorgement, and brain sagging [5]. MRI findings are variable, but the association between imaging data and clinical manifestations is not fully understood. There was no statistically significant difference in the incidence of abnormal brain MRI findings in SIH and PDPH patients, but more abnormalities were observed in SIH patients. We speculated that abnormal MRI findings were related to the interval between symptom onset and the MRI scans. Since the MRI findings of intracranial hypotension are compensatory reactions to extradural CSF leakage, patients with a long-standing history of intracranial hypotension are more likely to display atypical clinical and imaging findings [20]. Thus, a variety of disease severity and duration could lead to various combinations of abnormal MRI findings.

The most common brain MRI finding in patients with SIH was diffuse pachymeningeal enhancement, followed by subdural fluid collection. The order of brain MRI abnormalities in our study was similar to that reported previously [21]. In PDPH, brain MRI was thought to show normal findings because the procedures suddenly caused symptoms, but patients with PDPH also showed abnormal brain MRI findings in this study. Venous sinus distension (39.1%) was the most commonly observed abnormal MRI finding in the PDPH group, while the subdural fluid collection was observed in only one patient (4.3%). These findings are presumed to be because venous sinus dilatation occurs early in the intracranial hypotension process, and a chronic change in intracranial hypotension causes subdural fluid collection [22]. The vG/SS angle was significantly greater in the PDPH group than in SIH group. Severe transtentorial brain sagging causes stretching of the vG, which narrows the vG/SS angle. Since brain sagging also reflects a chronic change in intracranial hypotension, the vG/SS angle decreased in SIH, but not in PDPH patients [17].

EBP is the treatment of choice for patients who have not responded to conservative management for intracranial hypotension [1]. The efficacy of the first EBP for SIH is about 30% ~ 60%, and multiple EBPs are often required [1, 8, 9]. On the other hand, the efficacy of EBP in PDPH is much better, and the first EBP has a success rate of approximately 70 to 90% [23]. Even with accidental dural tears due to epidural catheterizations, the efficacy of response to EBP is better than that in SIH. This could be because, in PDPH, EBP is typically targeted right at or very close to the leakage site [24]. In contrast, the leakage site in SIH is mostly present at the levels above the lumbar spine where the EBPs are placed. Moreover, the dural defect in SIH is not a simple hole but occurs through three main mechanisms: meningeal diverticular lesions, ventral dural tears, and CSF-venous fistula [25]. Although targeted EBP in SIH may be more successful than blinded EBP [9], there is controversy over whether targeted or blind EBP is better for SIH treatment [26]. EBP in the lumbar area is safer than that in other areas and blinded EBP can be performed without other invasive or expensive tests to confirm the leakage site. It is argued that the goal of blind lumbar EBP is not to seal and repair the CSF leakage site, but to reverse the CSF pressure-venous pressure gradient within the spinal epidural space [27]. Ohtonari et al. reported that a large-volume EBP, using an intravenous catheter at a single lumbar entry point, provided complete CSF leak control in 14 SIH and 1 traumatic CSF leak patients with suspected multiple CSF leakage sites [28]. Even in our study, blinded EBP was safe and effective. The response rate of conservative management and up to two blinded EBP treatments was 90% in all SIH patients. The number of patients who underwent EBP three or more times was similar in the SIH (9.8%) and PDPH (9.1%) groups.

In the SIH group, EBP was performed more frequently in women, consistent with the results published in another study [29]. Under the influence of estrogen, swelling of the cerebral blood vessels in CSF hypovolemia is likely to be more pronounced in young women, leading to more severe headaches [30]. Moreover, there are reports that pain sensitivity may be higher in women [31], so EBP is more likely to be required by women. Although not statistically significant due to the small number of patients, SIH patients who underwent EBP showed more pachymeningeal enhancement and venous distension of the lateral sinus and quantitative signs that suggest brain sagging more frequently than patients received conservative management (supplemental Table 1).

Among PDPH patients, the EBP group showed a significantly longer interval between symptom onset and diagnosis than the conservative management group (supplemental Table 2). EBPs were likely performed earlier without conservative management in patients who already had symptoms for an extended period. The number of EBPs significantly differed according to the cause of PDPH. Among patients who developed PDPH by lumbar puncture, symptoms improved by only conservative management without EBP in most cases (13 out of 16), maybe because there was only a stab injury without significant damage to the dura mater. Among patients with PDPH caused by procedures other than a diagnostic lumbar puncture, such as epidural injection, EBP was frequently performed in 59.3% (16 of 27 patients); since epidural injection induces dural tears and spinal anesthesia involves an injection of drugs into the spinal cavity, these procedures are more invasive than a diagnostic lumbar puncture.

The present study had several limitations. First, the number of patients included in this study was small. If the number of patients was larger, more clinical characteristics and MR findings might have shown significant differences between SIH and PDPH patients. Second, there were fewer brain MRI studies in PDPH patients. Among SIH patients, 86.9% (53 of 61) underwent brain MRI, whereas, among PDPH patients, only 52.3% (23 of 44) underwent brain MRI. Third, this was a retrospective study, and a more extensive prospective study is needed to confirm these results.

Our study showed no difference in clinical characteristics between the SIH and PDPH groups. The vG/SS angle in the brain MRI, which reflects brain sagging, was significantly narrower in the SIH group than in the PDPH group. The success rate of EBP in SIH patients was lower than that of PDPH, and more patients with SIH received EBP treatments and needed multiple EBP treatments than patients with PDPH. Notably, in PDPH patients, the cause of PDPH was important. Most patients with PDPH due to lumbar puncture improved symptoms with conservative management without EBP. SIH and PDPH can be treated without significant complications if they are managed with appropriate investigation and treatment. It is a clinical decision to provide EBP to patients who have failed conservative treatment; therefore, information regarding clinical characteristics and MRI findings is necessary for treatment decisions.