Abstract
Objective: This is an analytic, interventional, cross sectional study to evaluate the risk factors of post-dural (post-lumbar) puncture headache (PDPH) and the validity of the diagnostic criteria for PDPH from the ICHD II.
Methods: Six-hundred-and-forty patients (332 non-pregnant women and 308 men) aged 8–65 years underwent spinal anesthesia with Quincke 25G or 27G needles in elective surgery.
Results: Forty-eight (7.5%) of the patients developed PDPH. The binary logistic regression analysis identified as risk factors: gender [11.1% female vs. 3.6% male, OR 2.25 (1.07–4.73); p = 0.03], age [11.0% 31–50 years of age vs. 4.2% others, OR 2.21 (1.12–4.36); p = 0.02], previous history of PDPH [26.4% positive vs. 6.2% negative, OR 4.30 (1.99–9.31); p < 0.01] and bevel orientation [16.1% perpendicular vs. 5.7% parallel, OR 2.16 (1.07–4.35); p = 0.03]. The period of latency between lumbar puncture and headache onset range from 6 to 72 hours and the duration from 3 to 15 days. In 34/48 (71%) patients with PDPH, at least one of the following was present: neck stiffness, tinnitus, hypoacusia, photophobia, or nausea.
Conclusion: In conclusion, 14/48 patients (29%) suffered none of the above-mentioned symptoms, indicating that a significant number of patients may suffer from PDPH in the absence of any symptoms apart from the headache itself. This suggests that a further analyses of existing studies should be made to determine if a criteria change may need consideration.
Keywords
Introduction
Post-dural (post-lumbar) puncture headache (PDPH) is a common and incapacitating complication following an intentional puncture of the dura-arachnoid, whether for the purposes of diagnosis, therapy, or spinal anesthesia, or inadvertently during an epidural procedure (1–5). The typical feature of PDPH is the postural character of the pain, that is when the individual assumes an upright position (e.g. sitting, standing). According to the diagnostic criteria of the International Headache Society (IHS) (6), the headache appears up to the fifth day after puncture and disappears spontaneously within a week, or up to 48 hours after an epidural blood patch. It is accompanied by at least one of the following symptoms: neck stiffness, tinnitus, hypoacusia, photophobia, and nausea. The IHS criteria are as follows (6):
Headache that worsens within 15 minutes after sitting or standing and improves within 15 minutes after lying down, with at least one of the following and fulfilling criteria C and D: 1. neck stiffness; 2. tinnitus; 3. hypoacusia; 4. photophobia; 5. nausea Dural puncture has been performed Headache develops within 5 days after dural puncture Headache resolves either:
a
spontaneously within 1 week within 48 hours after effective treatment of the spinal fluid leak (usually by an epidural blood patch)
Regarding patients and puncture techniques, the following factors may be mentioned as possible contributors to the development of PDPH: youth (7), female gender (8), large caliber-needles (9), needle tip type (cutting rather than non-cutting or atraumatic) (9), low-normal rather high (obese) body mass index, pregnancy and labor, history of recurrent headaches (10), the needle bevel perpendicular to the long axis of the spinal column (11), ambulation vs. recumbence after the post-dural puncture (12), and not reinserting of stylet vs. reinsertion (13). There is still debate about the angle of the needle in relation to the perforation of the dura mater resulting in a loss of cerebrospinal fluid (CSF) and PDPH (14,15). A previous history of PDPH being a risk factor for a new attack of PDPH was also reported (16,17).
The aims of the present study were to evaluate: 1) the influence of various factors relating to the patient and puncture technique on the development of PDPH; and 2) the validity of the diagnostic criteria for PDPH from the International Classification of Headache Disorders (6).
Methods
This is an analytic interventional, cross sectional study. Over a period of 12 months, we studied 640 patients (332 non-pregnant women and 308 men) between the ages of 8 and 65 years who had undergone spinal anesthesia with Quincke 25G (n = 239) and 27G (n = 401) needles in elective surgery (urological, orthopedic, abdominal, gynecological, vascular, and repair plastic surgery). Of the 640 patients, 342 had previously undergone spinal anesthesia and 53 of these had had PDPH.
In the operating room, all were monitored, sedated with midazolam and hydrated with a crystalloid solution. The spinal anesthetic was applied in the lumbar region, in the L2–L3, L3–L4, or L4–L5 intervertebral space, with the patient in the sitting position, and 0.5% bupivacaine, with or without glucose, was used as the local anesthetic. When the dura-arachnoid was punctured, information was collected regarding the variables relating to the puncture technique: caliber of the Quincke needle (25G or 27G), the orientation of the needle bevel to the long axis of the spinal column (parallel or perpendicular) and the angle of approach to the dura (median or paramedian).
In the postoperative period, all the patients were advised to sit up and walk about as soon as their clinical condition permitted. Postural headache, whether the pain appeared or worsened after the individual sat up or stood up, and whether it got better or disappeared after lying down, was investigated by a single researcher (JAA), by means of variable questions and daily visits to patients and/or through telephone contact up to the seventh day after the lumbar puncture. Patients with headache having the above-mentioned characteristics were also evaluated according to: 1) the period of latency between lumbar puncture and the onset of the headache; 2) how long the postural headache lasted; and 3) the presence of any of the following symptoms accompanying the postural headache: neck stiffness, tinnitus, hypoacusia (reduction of auditory sensitivity), photophobia or nausea.
Clinical treatment for the patients with PDPH was rest, anti-inflammatory non-hormonal analgesics, and 300 mg/day caffeine, taken orally (18).
A brain MRI (to disclose an intracranial complication) and an epidural blood patch were performed when the PDPH did not spontaneously disappear within a week or when the patient was severely impaired by the headache and/or accompanying symptoms.
Data are shown as mean, dispersion (standard deviation and range of variation) and frequency distribution. Occurrences of PDPH were estimated in percentages of the total sample, listed according to the patient variables (gender, age, previous history of PDPH) and puncture techniques (needle caliber, bevel orientation, angle of approach) and their respective categories. Binary logistic regression analysis was used to identify risk factors for the occurrence of PDPH. The variables with values in the analysis of p ≤ 0.20 were included in the final regression model (multivariate analysis) and those with p < 0.05 were considered as risk factors for PDPH.
Every patient or his or her legal representative gave informed written consent to participate in the study, which was approved by the ethics committee of the institutions involved.
Results
Physical and demographic characteristics of 48 patients with post-dural (post-lumbar) puncture headache.
Numbers represent the mean ± standard deviation (min–max).
Body mass index.
Incidence of post-dural (post-lumbar) puncture headache and regression binary logistic analysis of variables evaluated in this study.
The first level of bivariate regression analysis (raw odds ratio) indicates that the following had a statistically significant association with the occurrence of PDPH (p < 0.05): gender, age, previous history of PDPH, needle caliber, and bevel orientation. All the variables were included in the multivariate analysis (adjusted odds ratio) and the following maintained a statistically significant association: gender, age, previous history of PDPH, and orientation of the needle bevel. The values of the raw and adjusted odds ratios, the respective confidence levels, and the p values are shown in Table 2.
The risk of a new occurrence of PDPH was 4.3 times greater in patients with a previous history of PDPH [IC (1.99–9.31); p < 0.01]. All the patients who suffered a new attack of PDPH were women, although ten men had had a previous positive history. The interval between the previous attack and the present one ranged from 2 to 21 years (11.5 ± 7.4 years). Two women had three consecutive attacks of PDPH. No statistically significant associations between the angle of approach and PDPH were found.
Clinical features of the post-dural (post-lumbar) puncture headache in the 48 patients.
Discussion
We observed in this study that 48 (7.5%) patients presented postural headache characteristic of PDPH, and the independent risk factors for PDPH were female gender, age between 31 and 50 years, a previous history of PDPH and the orientation of the bevel perpendicular to the long axis of the spinal column at the time of the dura mater/arachnoid puncture.
Women presented a risk of PDPH 2.25 times greater than that of men, a result similar to that of Wu and colleagues (8), who, in a meta-analytical study of non-pregnant women, found that the risk of PDPH was twice as great as that of men, irrespective of age, needle caliber or design of the bevel. Possible explanations lie in the physiological, anatomical, social, behavioral characteristics peculiar to women, as well as their perception of pain. High levels of estrogen in women can influence the tone of the cerebral vessels, thus increasing the vascular distension response to CSF hypotension (19), particularly in the premenopause phase (20), which is the age range that showed the greatest risk of PDPH in the present study. In this phenomenon, excitatory neurotransmitters are released (21) and the perivascular sensory nerves that surround the intracranial veins and sinuses are stimulated, supporting the hypothesis of intracranial venous vasodilatation as the cause of headache, as does the diminution of pain with analgesics (e.g. caffeine) having cerebral vasoconstrictor activity (22). Furthermore, women seem to process nociceptive information differently from men, showing greater sensitivity to painful stimulation, which facilitates the central sensitization process, as has been shown in neuroimaging studies (23,24).
A greater risk of PDPH in the 31–50 years age range was also found by Wadud and colleagues in a study that compared patients under and over 50 years of age (25). Other studies have found a higher incidence of PDPH between the ages of 20 and 30 years (16,26,27). Therefore, it is unquestionable that the incidence of PDPH is higher in adults, decreasing with increasing age. Although the physiopathology is not clear, in theory three factors could prevent individuals over 50 years from developing PDPH: (a) the reduced elasticity of the dura mater, which makes it more difficult for CSF to leak through the puncture hole, (b) a weaker reaction of the cerebral vessels to CSF hypotension, and (c) a reduced vertebral extradural space allowing a small amount of CSF accumulation, thereby arresting the leak of CSF from the subarachnoid space (increased extradural resistance) (7,28). PDPH is not common in children, which could be explained by the fact that there is less physiological pressure from the CSF, in addition to a low hydrostatic pressure in the lumbar region when these patients sit up, compared with adults (29,30).
There are few published studies evaluating the risk of a new PDPH attack in patients with a previous history. Lybercker et al. (16) studied 1021 spinal anesthesias: 117 of the patients had previously undergone spinal anesthesia and three had suffered from PDPH. These three patients were given a second dural puncture, and two of them (66%) suffered a new attack. Of the 114 who had no previous history, only three (2.8%) developed PDPH (16). In a controlled study by Amorim and Valença (17), involving 258 patients who had a previous history of dural puncture for spinal anesthesia, 42 patients had had a previous history of PDPH, while 216 had no previous history. When these patients were given a second dural puncture, the prevalence of PDPH was significantly greater in the group with a previous positive history [8/42 (19.0%) vs. 15/216 (6.9%); p = 0.0118)]. Women had a greater risk of a recurring PDPH (17). In the present study, which involved a larger number of patients, we have corroborated this hypothesis and believe that the physiological characteristics peculiar to females, including the thickness of the dura-arachnoid, which facilitates CSF leakage when the dura is thinner, could possibly account for the susceptibility of women to consecutive episodes of PDPH (31).
Recently, we demonstrated (32) that, after perforating a human cadaveric dura mater using a model of a dural sac with the help of an acrylic column with dural attachment mimicking an in vivo situation (40 cm H2O pressure at the level of puncture), the outflow of liquid was higher using female-derived dura mater fragments than male-derived ones. In addition, after perforation of the dura mater the initial liquid outflow was highly variable between dura mater specimens [ranging from 18 ml/10 min to no outflow at all, using a 27-gauge Quincke needle], even when different fragments of the same cadaver donor were used, thereby explaining why only some individuals developed PDPH. Another noteworthy point is the observation that during the 60-minute experiment the liquid outflow decreased with time and some of the perforated fragments show a spontaneous arrest of the liquid outflow. In a few of the tested dura mater fragments we did not observe any loss of liquid after the perforation of the dura mater by the insertion and removal of the needle. This demonstrated that the dura mater has an intrinsic elastic mechanism enabling it to restore or occlude the orifice produced by the needle and that this characteristic is variable between different tested specimens and with the gender of the cadaver donor (32).
With respect to puncture technique, we expected that the caliber of the needle would continue to be a risk factor in a multivariate analysis, although there would probably be a series of factors influencing this result, such as the smaller number of punctures with the 25G needle, the proximity of calibers between the 25G and 27G needles, and the fact that both needles have a similar tip bevel design. Lybecker et al. (16), in a study to evaluate multiple factors in predicting PDPH resulting from the use of 22G, 25G and 26G cutting needles, also found in regression analysis that caliber was not a predictive factor for PDPH; however, the perpendicular orientation of the bevel was a predictive factor, as was also found in this study. Regarding the unanticipated outcome of null effect of needle gauge (caliber) as risk factor of PDPH presented here and by other authors (16), multiple meta-analyses have consistently demonstrated the convincing effect that the needle caliber has on development (risk) of PDPH (9,10,33).
Richman et al. (11), in a meta-analysis study, evaluated the influence of Quincke and Tuohy needles (cutting bevel) on the incidence of PDPH in adult patients. They also demonstrated that a bevel orientation parallel to the long axis of the spinal column significantly lowered the incidence of PDPH, when compared with the perpendicular [10.9% vs. 25.8%; OR 0.29 (95% IC = 0.17–0.50)] (11). In this regard, it has been postulated that the arachnoid may be at least as important as the dura mater, and indeed perhaps more so, in the genesis of PDPH. Kempen and Moeck (34), in an experimental study, found that, when puncture was made with a parallel orientation of the needle bevel, the layers of dura mater and arachnoid overlap and this may reduce CSF leakage. Zetlaoui (35) suggested that variations in the diameter of the puncture hole in the dura mater are due to the movements and pricks made in the dural sac, which is immobile in the vertebral column. When the bevel is parallel to the neural axis, the prick, which opens up when the patient sits up or stands, tends to close. With the bevel perpendicular, the hole is enlarged and there is a subsequent greater loss of CSF (35).
As for the angle of insertion of the needle, an in vitro study using a model of human dura mater demonstrated a smaller loss of CSF when the needle was inserted using the paramedian approach (0.3 ± 0.4 ml/min), whereas when the median approach was used the loss of CSF was greater (3.3 ± 1.6 ml/min) (14). One possible explanation would be that the paramedian approach decreases the loss of CSF resulting from perforation of the dura mater and the arachnoid at different angles, producing a valvular mechanism that prevents a greater CSF flow to the epidural space. In the present study, no significant association between the angle of approach and the incidence of PDPH was found. Imbelloni et al. (36) likewise found no association between the angle of approach of the needle and the incidence of PDPH in a study with pencil tipped needles and a cutting bevel, caliber 25G and 27G.
In the evaluation of the clinical features, when we compared our results with the diagnostic criteria of PDPH defined by the IHS (6), the latent period established for the onset of PDPH (up to 5 days after puncture) proved to be satisfactory. In all our 48 patients with postural headache, the onset of the pain occurred up to 72 hours after puncture. This finding has been similarly demonstrated in other studies (26,27). As for the period established for the disappearance of the symptoms, one patient in this study failed to fulfill the diagnostic criteria because her headache lasted for more than 7 days, even after an epidural blood patch had been applied. As has been observed in most of the patients, the duration of PDPH is 5 days and in more than 95% of the cases the headache resolves spontaneously within 1 week. However, it is possible, albeit unusual, for the postural headache to last more than 1 week.
When considering the duration of headache, the original study reported by Vandam and Dripps (26) is the largest and most definitive. In a study of 10,098 spinal anesthesias involving 8460 patients, they followed the patients for 2 weeks after the dural puncture and then performed a follow-up at 6 months (26). The needles used were Quincke 16G to 24G. They reported that just 72% of PDPH resolved within 7 days (26). Vilming and Kloster (37) studied 289 who had undergone diagnostic lumbar puncture with needles 20/22G. They reported 36.8% of PDPH: all the cases appeared up to 4 days after puncture and the median duration was 6 days (range 1–29 days) (37). Klepstad (38) reported a case of a healthy 20-year-old man who suffered from persistent postural headache for a year after a spinal puncture. This patient was relieved only after being given an epidural blood patch (38). Although this situation does occasionally arise, when the headache persists for over 1 week, PDPH is unlikely to be the cause and a magnetic resonance of the skull should be recommended to rule out other intracranial complications, such as subdural hematoma (39), which may occur after a lumbar puncture.
The literature reports a number of patients with persistent and untreated PDPH who subsequently developed intracranial subdural hematoma (39). In situations such as these, any delay in diagnosis and treatment increases the chance of complications and, in rare cases, the syndrome of cerebrospinal fluid hypotension may even be the cause of death (40,41).
Finally, we observed that nearly one third of the patients who presented with postural headache would not have been included as PDPH sufferers according to the IHS diagnostic criteria. In 14 of the 48 patients with postural headache the usual accompanying symptoms were absent. This result indicated that a significant number of patients may suffer PDPH without any accompanying symptoms. A similar result was found in a prospective study by Vilming and Kloster (42) with 239 patients who had undergone diagnostic lumbar puncture, where 15% of the patients with PDPH did not complain of any symptoms accompanying the postural headache.
Nausea was the most prevalent accompanying symptom in the present study, as reported in previous studies (37,42). The explanation for the vestibular and cochlear symptoms is that the low pressure of the CSF is transmitted through a patent cochlear aqueduct to the inner ear (43). The space of the cochlear tube may vary greatly among individuals and even those with a large tube may have it completely obstructed by the arachnoid tissue, leading to a lack of communication between the two compartments (43). Because the permeability between the subarachnoid space and the inner ear varies so much between individuals, and changes in the pressure of the CSF flowing in the subarachnoid space may be communicated to the structure of the inner ear, we can therefore explain why the vestibular-cochlear symptoms may be absent and, when present, appear to correlate with the intensity of PDPH.
Visual disturbances are caused by dysfunction of the extraocular muscles from transient paralysis of the motor nerves of the eye (III, IV, and IV cranial nerves). The abducens nerve is the nerve most frequently affected, in part owing to its long intracranial course (44).
Regarding the treatment of patients with PDPH, some authors declare that there is no valid pharmacological rationale for caffeine as an antinociceptive agent for PDPH (45). The clinical trials are few in number, small in sample size, methodologically weak or flawed, and either demonstrate no effectiveness, contradictory and conflicting results, or invalid answers (45).
A recent study (18), however, concluded that caffeine has shown some effectiveness for treating PDPH, reducing the proportion of participants with PDPH persistence and the number requiring supplementary interventions when compared with placebo. Pain severity scores were also lower with gabapentin, theophylline and hydrocortisone. Nevertheless, the authors emphasized that their conclusions should be interpreted with caution because of the lack of information needed to allow correct appraisal of risk of bias, small sample sizes and the limited generalisability, as most participants were young post-partum women (18).
In the present study we concluded that several factors seem to predispose a patient to develop PDPH after spinal anesthesia. The identification of factors that predict the likelihood of PDPH is important so that measures can be taken to minimize this painful complication resulting from a dural puncture. In addition, the ICHD II (6) diagnosis criteria were insufficient for classifying all patients with PDPH.
It may be premature to propose IHS criteria changes based on the present study, but we propose further analyses of existing studies to determine whether a criteria change may need consideration, taking into account that: (i) the postural headache may or may not be accompanied by vestibular, visual and auditory symptoms, or neck stiffness; and (ii) when the headache is typically postural but persists longer than 1 week, with an MRI scan ruling out the presence of intracranial lesions and a CSF examination confirming hypotension and the absence of infection or hemorrhage, it is still a PDPH.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
a
In 95% of cases this is so. When headache persists, causation is in doubt.
