Headaches, which are very common in the general population (1,2), are also a common symptom of raised intracranial pressure (ICP) (3). Many patients with new-onset headaches, especially obese young women, undergo lumbar puncture (LP) for possible elevated ICP. Such patients may be diagnosed with idiopathic intracranial hypertension (IIH) based on a single elevated opening pressure (OP), even without an ophthalmoscopic examination (4). Fisayo et al found that 39.5% of patients referred for a diagnosis of IIH were ultimately determined not to have IIH. Among these patients, 79% had an LP with 44% of them having an “elevated” OP of ≥25 mm H2O. In a nonspecialists' hand, dependence on a single elevated LP OP creates misdiagnoses that can lead to more invasive tests and treatments, including cerebrospinal fluid (CSF) shunting (4,5). The aim of this study was to evaluate consecutive LP OPs to determine the expected potential OP variability that can occur between LPs, which could contribute to spuriously elevated LP OPs.
In this retrospective cohort study, medical records of patients who had at least 2 LPs with OPs within 30 days of each other at Mayo Clinic, Rochester, MN, from January 1, 2001, through June 1, 2016, were reviewed. This study was approved by the institutional review board of Mayo Clinic, Rochester, MN, and it conforms to the requirements of the US Health Insurance Portability and Privacy Act.
A total of 464 patients' medical records were reviewed. The excluded diagnoses were those related to central nervous system inflammation or infection that may cause ICP fluctuations (e.g., meningitis), obstructive hydrocephalus, IIH, papilledema, arteriovenous fistula, hemorrhagic stroke, low CSF pressure/leak, and space-occupying lesions shifting intracranial structures or causing hydrocephalus. Patients who had undergone medical or surgical treatment that could alter ICP between the LPs were also excluded.
After the above exclusions, the diagnoses included in the study came under the categories of normal pressure hydrocephalus, parkinsonism, dementia, headache, ischemic stroke, nonspecific gait disturbance, neurosarcoidosis, seizures, small nonobstructing neoplasm, amyotrophic lateral sclerosis, encephalopathy, neuropathy, and multiple sclerosis. There were 21 patients whose final diagnoses were indeterminate at last follow-up; they were included since they did not meet any of the exclusionary criteria.
Age, gender, body mass index (BMI), race, LP OP, and diagnoses of headaches, anxiety, or depression before the first LP were recorded.
The LP OP variability was evaluated using Bland and Altman, and coefficient of repeatability (CR) (6). CR, which is an expression of 2 SDs, provides the expected absolute difference between LP OPs with a 95% probability. A higher CR indicates higher variability and provides the absolute difference expected between the LP OPs with only 5% of instances expected to have a higher difference between 2 LP OPs. These values were compared using a standard F-test to compare variances (7). The difference between consecutive LP OPs was compared with paired tests. Pearson correlation coefficients were used to test the association between continuous measures.
Of the 148 patients who underwent at least 2 LPs within 30 days of each other, 60.8% were men and 88.5% were Caucasians (Table 1). The mean age was 63.5 ± 15.5 years, with a mean BMI of 28.0 ± 6.0 kg/m2. Mean OP of the first LP was 149 ± 51 mm H2O and for the second LP was 148 ± 48 mm H2O (P = 0.78), with a mean difference of 1 mm H2O and a mean of 10 ± 9 days between LP OPs. Mean OPs from consecutive LPs among women were 144.4 mm H2O and 141.2 mm H2O, compared with 152.0 mm H2O (P = 0.38) and 152.5 mm H2O (P = 0.17) in men.
TABLE 1. -
Demographics of 148 patients with ≥2 LPs within 30 days
|Age at LP #1 in years, mean (range)
|Body mass index (BMI) at LP #1 (kg/m2), mean (range)
|Missing BMI data
BMI, body mass index; LP, lumbar puncture.
Overall CR was 86.4 (Table 2). While initial LP OP less than 200 mm H2O had a CR of 74.6, OP greater than 200 mm H2O had a CR of 111.4 (P = 0.006).
TABLE 2. -
Coefficient of repeatability (CR) for initial LP OPs showing higher initial LP OPs was associated with greater variability
||Coefficient of Repeatability
|LP #1 OP ≤200 mm H2O
|LP #1 OP >200 mm H2O
*P value compares the coefficient repeatability of LP ≤ 200 and >200 mm H2O.
LP, lumbar puncture; OP, opening pressure.
Mean LP OPs were mapped against differences between OPs in a Bland–Altman plot (Fig. 1). Age was not found to have a significant correlation with either initial LP OP (r = 0.04, P = 0.59) or the difference between OPs (r = 0.07, P = 0.42). There was a significant correlation between BMI and initial LP OP (r = 0.39, P < 0.001), but not between BMI and the differences between OPs (r = 0.11, P = 0.24). No association was found between LP OP and the time of day at which the LP occurred.
Comparing the CRs of LP OPs in patients with headache, anxiety, or depression against those without these diagnoses, there was significantly greater variability among those with headaches (CR 118.1, P = 0.002) and anxiety (CR 112.9, P = 0.03), but not depression (CR 75.5, P = 0.52) (Table 3).
TABLE 3. -
Coefficient of repeatability (CR) for patients with headache, anxiety, or depression
||Coefficient of Repeatability
*Diagnoses were rendered before lumbar puncture #1.
Since the first LP was described in the 19th century, the procedure has gained popularity as both a diagnostic and therapeutic tool (8,9). LP OP has been adopted widely as a proxy for ICP, although its use has not been fully standardized and even the reference ranges for both pediatric (10–12) and adult (13–15) patients continue to be debated. Nevertheless, some diagnostic criteria have incorporated OP cutoffs, including the criteria for spontaneous intracranial hypotension (16) and the modified Dandy criteria for IIH (17). Our study reveals considerable variability between consecutive LP OPs, by almost 75 mm H2O when the initial OP is <200 mm H2O and by over 110 mm H2O when the initial OP is >200 mm H2O. Such variability may make it difficult for physicians to determine whether a patient truly has normal or sustained elevated ICP based on a single OP measurement.
Several factors contribute to the uncertainty regarding normal vs pathologic OP ranges, including age, gender, BMI, patient and manometer positioning, Valsalva maneuvers, anesthesia, needle size, and time of day (13,18,19). LP OPs have previously been reported to increase with age in children until 8 years (19); decline with age in adults after 50 years (20); be higher in men (13); increase with BMI (15); increase in the flexed, seated, or prone positions (13,21–24); increase with volitional Valsalva maneuvers (25); increase by 30–50 mm H2O in children with sedation (11,26); be slower to equilibrate with smaller-bore and longer needles (27); and fluctuate over the course of a day (28,29).
In this study, initial LP OP was significantly associated with BMI but not with age, and the difference between consecutive LP OPs was not associated with either trait. This was in contrast to previous findings of a gradual decline in ICP with increasing age after 50 years (20), an association this study may have been underpowered to find. There was no significant difference in LP OP between men and women, and no diurnal pattern was found among the OPs. Greater variability in OP was significantly associated with diagnoses of headache and anxiety, but not depression. Because Valsalva can increase OP by as much as 310 mm H2O (25), it is possible that patients with headache and/or anxiety may be more prone to this variability.
Almost all the patients in this study had pathologic indications for undergoing multiple LPs within a month, albeit not pathologies expected to affect OP. These data thus do not represent what might be expected in a “normal” population. Since healthy subjects tend not to undergo LP, however, these patients undergoing LPs for common reasons (e.g., headache, altered mental status, gait disturbance, and seizures) perhaps represent precisely the desired reference group. Because of the retrospective nature of the study, complete assessment of all variables could not be performed, with 13.5% missing BMI and some missing certain details such as needle gauge. Finally, the study population was largely Caucasians. The effect of ethnicity on LP OP, if any, is unknown, and so, our results may not be generalizable to other populations.
Our data suggest that there is significant variability in consecutive LP OPs, especially at pressures greater than 200 mm H2O. OPs that are not consistent with the patients' clinical signs and symptoms should therefore be interpreted with caution, and repeating the LP may be reasonable if the OP is essential for diagnosis.
STATEMENT OF AUTHORSHIP
Category 1: a. Conception and design: J. J. Chen, K. P. Kilgore, and D. O. Hodge; b. Acquisition of data: K. P. Kilgore and J. J. Chen; c. Analysis and interpretation of data: J. J. Chen, K. P. Kilgore, and D. O. Hodge. Category 2: a. Drafting the manuscript: J. J. Chen and K. P. Kilgore; b. Revising it for intellectual content: J. J. Chen, K. P. Kilgore, M. T. Bhatti, and J. K. Cutsforth-Gregory. Category 3: a. Final approval of the completed manuscript: J. J. Chen, K. P. Kilgore, T. Bhatti, J. K. Cutsforth-Gregory, and D. O. Hodge.
1. Rasmussen BK, Jensen R, Schroll M, Olesen J. Epidemiology of headache in a general population-a prevalence study. J Clin Epidemiol. 1991;44:1147–1157.
2. Hagen K, Zwart JA, Vatten L, Stovner LJ, Bovim G. Prevalence of migraine and non-migrainous headache-head-HUNT, a large population-based study. Cephalalgia. 2000;20:900–906.
3. Wall M, George D. Idiopathic intracranial hypertension. A prospective study of 50 patients. Brain. 1991;114:155–180.
4. Fisayo A, Bruce BB, Newman NJ, Biousse V. Overdiagnosis of idiopathic intracranial hypertension. Neurology. 2016;86:341–350.
5. Xu DS, Hlubek RJ, Mulholland CB, Knievel KL, Smith KA, Nakaji P. Use of intracranial pressure monitoring frequently refutes diagnosis of idiopathic intracranial hypertension. World Neurosurg. 2017;104:167–170.
6. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1:307–310.
7. Mendenhall W. Introduction to Probability and Statistics. 7th edition. Boston, MA: Duxbury Press, 1986:454.
8. Doherty CM, Forbes RB. Diagnostic lumbar puncture. Ulster Med J. 2014;83:93–102.
9. Frederiks JA, Koehler PJ. The first lumbar puncture. J Hist Neurosci. 1997;6:147–153.
10. Ellis R III. Lumbar cerebrospinal fluid opening pressure measured in a flexed lateral decubitus position in children. Pediatrics. 1994;93:622–623.
11. Avery RA. Interpretation of lumbar puncture opening pressure measurements in children. J Neuroophthalmol. 2014;34:284–287.
12. Cartwright C, Igbaseimokumo U. Lumbar puncture opening pressure is not a reliable measure of intracranial pressure in children. J Child Neurol. 2015;30:170–173.
13. Lee SC, Lueck CJ. Cerebrospinal fluid pressure in adults. J Neuroophthalmol. 2014;34:278–283.
14. Whiteley W, Al-Shahi R, Warlow CP, Zeidler M, Lueck CJ. CSF opening pressure: reference interval and the effect of body mass index. Neurology. 2006;67:1690–1691.
15. Bono F, Quattrone A. CSF opening pressure: reference interval and the effect of body mass index. Neurology. 2007;68:1439; author reply 1439–1440.
16. Schievink WI, Maya MM, Louy C, Moser FG, Tourje J. Diagnostic criteria for spontaneous spinal CSF leaks and intracranial hypotension. AJNR Am J Neuroradiol. 2008;29:853–856.
17. Friedman DI, Liu GT, Digre KB. Revised diagnostic criteria for the pseudotumor cerebri syndrome in adults and children. Neurology. 2013;81:1159–1165.
18. Ducros A, Biousse V. Headache arising from idiopathic changes in CSF pressure. Lancet Neurol. 2015;14:655–668.
19. Avery RA, Shah SS, Licht DJ, Seiden JA, Huh JW, Boswinkel J, Ruppe MD, Chew A, Mistry RD, Liu GT. Reference range for cerebrospinal fluid opening pressure in children. N Engl J Med. 2010;363:891–893.
20. Fleischman D, Berdahl JP, Zaydiarova J, Stinnett S, Fautsch MP, Allingham RR. Cerebrospinal fluid pressure decreases with older age. PLoS One. 2012;7:e52664.
21. Avery RA, Mistry RD, Shah SS, Boswinkel J, Huh JW, Ruppe MD, Borasino S, Liht DJ, Seiden JA, Liu GT. Patient position during lumbar puncture has no meaningful effect on cerebrospinal fluid opening pressure in children. J Child Neurol. 2010;25:616–619.
22. Schwartz KM, Luetmer PH, Hunt CH, Kotsenas AL, Diehn FE, Eckel LJ, Black DF, Lehman VT, Lindell EP. Position-related variability of CSF opening pressure measurements. AJNR Am J Neuroradiol. 2013;34:904–907.
23. Dinsmore J, Bacon RC, Hollway TE. The effect of increasing degrees of spinal flexion on cerebrospinal fluid pressure. Anaesthesia. 1998;53:431–434.
24. Abbrescia KL, Brabson TA, Dalsey WC, Kelly JJ, Kaplan JL, Young TM, Jenkins D, Chu J, Emery MS. The effect of lower-extremity position on cerebrospinal fluid pressures. Acad Emerg Med. 2001;8:8–12.
25. Neville L, Egan RA. Frequency and amplitude of elevation of cerebrospinal fluid resting pressure by the Valsalva maneuver. Can J Ophthalmol. 2005;40:775–777.
26. Avery RA. Reference range of cerebrospinal fluid opening pressure in children: historical overview and current data. Neuropediatrics. 2014;45:206–211.
27. Bellamkonda VR, Wright TC, Lohse CM, Keaveny VR, Funk EC, Olson MD, Laack TA. Effect of spinal needle characteristics on measurement of spinal canal opening pressure. Am J Emerg Med. 2017;35:769–772.
28. Gucer G, Viernstein L. Long-term intracranial pressure recording in the management of pseudotumor cerebri. J Neurosurg. 1978;49:256–263.
29. Nornes H, Rootwelt K, Sjaastad O. Normal pressure hydrocephalus. Long-term intracranial pressure recording. Eur Neurol. 1973;9:261–274.