Greater than expected prevalence of pseudotumor cerebri: a prospective study

Transcription

1 Surgery for Obesity and Related Diseases 9 (2013) Original article Greater than expected prevalence of pseudotumor cerebri: a prospective study Isam N. Hamdallah, M.D., Hazem N. Shamseddeen, M.D., Jorge L. Zelada Getty, M.D., William Smith, Mohamed R. Ali, M.D.* Department of Surgery, University of California, Davis, Sacramento, California Received May 1, 2011; accepted November 8, 2011 Abstract Keywords: Background: The overall incidence of pseudotumor cerebri (PTC) has been estimated at 1 per 100,000 in the general population, with an increase to 19 per 100,000 among overweight patients. About 25% of affected patients are asymptomatic until they present with vision loss. We hypothesized that PTC would be highly prevalent among obese patients seeking bariatric surgery. The setting of our study was a university hospital. Methods: During a 2-year period, clinical data were collected from candidates for bariatric surgery. A group of the study population was concurrently screened for papilledema using fundus imaging. All images were reviewed by a single neuro-ophthalmologist. All patients with abnormal images were referred for neuro-ophthalmic evaluation. Results: The imaging group (78% women) had a mean age of years, and the mean body mass index of kg/m 2. High definition nonmydriatic fundus imaging was normal in 489 patients (91.9%) and abnormal in 43 patients (8.9%). The subsequent evaluation by the neuro-ophthalmologist revealed ocular abnormalities other than optic disc edema in 27 patients (5.1%) and normal findings in 7 patients (1.3%). Five patients (.9%) declined additional evaluation. Four patients (.8%) had confirmed optic disc edema and normal brain magnetic resonance imaging findings. Of these patients, 3 underwent lumbar puncture, which confirmed the diagnosis of PTC. Conclusion: We identified a greater prevalence of PTC overall (.65%) and in the imaging group (.9%) than previously reported or would have been predicted from the current data. Although routine funduscopic examination is of limited utility, these findings warrant additional investigation into the utility of, and optimal method for, screening morbidly obese patients for this co-morbidity. (Surg Obes Relat Dis 2013;9:77 82.) 2013 American Society for Metabolic and Bariatric Surgery. All rights reserved. Idiopathic intracranial hypertension; Pseudotumor cerebri; Bariatric surgery; Co-morbidity; Morbid obesity Obese individuals have a greater risk of developing a multitude of medical conditions, including diabetes mellitus, cardiovascular diseases, sleep apnea, and dyslipidemia [1]. Although these conditions are common among obese patients, pseudotumor cerebri (PTC), also known as idiopathic intracranial hypertension, is also associated with obesity and can lead to loss of vision [2]. *Correspondence: Mohamed R. Ali, M.D., Minimally Invasive and Robotic Surgery, Department of Surgery, University of California, Davis, 2221 Stockton Boulevard, Cypress Building, Sacramento, CA mohamed.ali@ucdmc.ucdavis.edu. PTC is a clinical syndrome in which the cerebrospinal fluid pressure is elevated in the absence of space-occupying lesions, vascular lesions, or enlargement of the cerebral ventricles and for which no causative factor can be identified [3]. Patients can present with the following symptoms: headaches, tinnitus, blurring of vision, or diplopia. Although the diagnostic criteria for PTC have been modified to reflect new imaging and diagnostic modalities [4 6], the etiology of this disease remains poorly understood [6,7]. PTC is associated with significant clinical sequelae and can lead to permanent vision loss [2] /13/$ see front matter 2013 American Society for Metabolic and Bariatric Surgery. All rights reserved. doi: /j.soard

2 78 I. N. Hamdallah et al. / Surgery for Obesity and Related Diseases 9 (2013) The overall incidence of PTC has been estimated at 1 2/100,000 in the general population, with an increase to 19/100,000 among overweight patients [8,9]. Although morbidly obese patients are considered at high risk of this co-morbidity, the data regarding the prevalence of PTC in this patient population are limited. We hypothesized that PTC would be highly prevalent among morbidly obese individuals seeking bariatric surgery. Methods From February 2008 to February 2010, 1084 patients were evaluated for bariatric surgery. The demographic data, anthropomorphic data, and co-morbidity data were prospectively collected for all patients. The co-morbidity severity was determined for all patients using the scoring system for the Bariatric Outcomes Longitudinal Database, a modification of the Assessment of Obesity-Related Co-morbidities scale [10]. This scale assigns a score of 0 5 for each co-morbidity according to the severity. Although the Bariatric Outcomes Longitudinal Database data were collected for many co-morbidities, the present study focused on the major metabolic conditions that might be related to PTC, including diabetes mellitus (DM), hypertension (HTN), dyslipidemia (DYS), sleep apnea syndrome (SAS), polycystic ovarian syndrome (PCOS), and PTC (Table 1). Addition- Table 1 Bariatric Outcomes Longitudinal Database/assessment of obesity-related co-morbidity severity scale Co-morbidity Score Description Diabetes mellitus Hypertension Polycystic ovarian syndrome Sleep apnea syndrome Dyslipidemia Pseudotumor cerebri 0 No symptoms or evidence of diabetes 1 Elevated fasting glucose 2 Diabetes, controlled with oral medication 3 Diabetes, controlled with insulin 4 Diabetes, controlled with insulin and oral medication 5 Diabetes, with severe complications (retinopathy, neuropathy, renal failure, blindness) 0 No history of hypertension 1 Borderline, no medication 2 Diagnosis of hypertension, no medication 3 Treatment with single medication 4 Treatment with multiple medications 5 Poorly controlled by medications, organ damage or dysfunction 0 No history of polycystic ovarian syndrome 1 Symptoms of PCOS, no treatment 2 OCPs or antiandrogen treatment 3 Metformin or TZD 4 Combination therapy 5 Infertility 0 No symptoms or evidence of SAS 1 SAS symptoms (negative sleep study or not done) 2 SAS diagnosis by sleep study (no oral appliance) 3 SAS requiring oral appliance, such as CPAP 4 SAS with significant hypoxia or oxygen dependent 5 SAS with complications (pulmonary hypertension, etc.) 0 Not present 1 Present, no treatment required 2 Controlled with lifestyle change, including step 1 or step 2 diet 3 Controlled with single medication 4 Controlled with multiple medications 5 Not controlled 0 No symptoms of pseudotumor cerebri 1 Headaches with dizziness, nausea, and/or pain behind the eyes, no visual symptoms 2 Headaches with visual symptoms and/or controlled With diuretics 3 Patient has had MRI to confirm PTC, is well-controlled with oral diuretics 4 Patient is well controlled with stronger medications 5 Patient requires narcotics or has had (or needs) surgical intervention PCOS polycystic ovarian syndrome; OCP oral contraceptive; TZD thiazolidinediones; SAS sleep apnea syndrome; PTC pseudotumor cerebri.

3 Prevalence of Pseudotumor Cerebri / Surgery for Obesity and Related Diseases 9 (2013) ally, all patients were offered the opportunity to undergo funduscopic imaging to screen for optic disc edema. Fundus images were captured with a nonmydriatic camera (Nidek/Gamagori, Aichi, Japan), which does not require dilation of the pupil to image the fundus. The images captured by the camera supersede those that can be obtained by a direct ophthalmoscope and allow detection of many funduscopic abnormalities, including slight disc edema. All images were obtained by the same technician who was specifically trained in the procedure. The images were then reviewed under the supervision of a single neuro-ophthalmologist. All ophthalmologic abnormalities were recorded, and all patients with major abnormalities were referred for additional clinical evaluation. Optic disc edema, the ophthalmologic abnormality associated with PTC, was graded using the Frisen scale [2]. This numeric grading system assigns a score of 0 to 5 as follows: 0, no edema; 1, very early edema; 2, early edema; 3, moderate edema; 4, marked edema; and 5, severe edema. Stage 1 manifests as excessive (in relation to the disc diameter) blurring of the nasal border of the optic disc, with disruption of the normal radial arrangement of nerve fiber bundles. The remaining stages (2 5) are associated with additional findings (temporal blur, peripheral total obscuration of major vessel segments by axons, central total obscuration of major vessel segments, and smooth dome). Any patient with a Frisen score of 1 (indicating the presence of optic disc edema) was evaluated by neuro-ophthalmology to confirm the presence of optic disc edema. Once optic disc edema was verified, the patient was neurologically evaluated with magnetic resonance imaging to exclude intracranial pathologic features and then underwent lumbar puncture. The present study was approved by the institutional review board. The data are reported as the mean standard deviation. Analysis of variance was used to compare the mean values of multiple groups of continuous variables. Analysis of discrete variables was accomplished using Fisher s exact test. Statistical significance was set at.05 for all analyses. Assuming the reported incidence of idiopathic intracranial hypertension in obese patients (19/100,000), our study was adequately powered to detect an absolute deviation from the overall population mean of.02% with 95% confidence. Results Of the 1084 patients evaluated for bariatric surgery, 568 (52.4%) underwent funduscopic imaging. Funduscopic images were successfully captured in 532 patients (imaging group). The remaining 36 patients had inadequate funduscopic imaging and were included with the nonimaging group. The imaging group was demographically representative of the entire patient population (Table 2). The 532 patients were predominately women (78%), with mean age of years and mean body mass index of kg/m 2. The vast majority of the imaging group (n 489, 91.9%) had completely normal funduscopic images. Patients whose images exhibited ocular abnormalities other than optic disc edema (n 27) were treated by the ophthalmology department as appropriate. These patients exhibited hypoplastic optic disc, asteroid hyalosis, optic nerve drusen, staphylomatous nerve, Hollenhorst plaque, proliferative diabetic neuropathy, chorioretinal scar, epiretinal membrane, dotblot hemorrhage and exudate, choroidal nevus, macular drusen, and glaucomatous nerve. Sixteen patients had images suspicious for optic disc edema. Five patients were not evaluated by the neuro-ophthalmologist, because they declined additional evaluation, and 7 patients were found not to have optic disc edema on additional clinical assessment. Evaluation by neuro-ophthalmology confirmed optic disc edema in 4 patients who subsequently underwent magnetic resonance imaging to exclude other causes for the edema. All 4 patients had early-stage optic disc edema (Frisen stage 1), no visual field loss, and normal visual acuity. Three of these patients then underwent lumbar puncture by interventional radiology to confirm PTC. The fourth patient declined lumbar puncture and chose to proceed with bariatric surgery. Thus, the diagnosis of PTC was not confirmed in that patient. The co-morbidity profiles of all patients are detailed in Table 3. These data indicate that the imaging group displayed metabolic derangements (hypothesized to predispose to PTC) similar to those of the nonimaging group and was a representative sample of the overall study population. Furthermore, the metabolic derangements associated with morbid obesity and hypothesized to coexist with PTC are prevalent within the imaging group (DM, 31.6%; SAS, 40.8%; HTN, 56.4%; PCOS, 4.9%; and DYS, 49.6%; Table 3). Overall, 4 patients (.4%) presented to our program with an established diagnosis of PTC (2 patients with a score of 4 in the imaging group and 2 patients with a score of 5 in the nonimaging group) and were being actively treated (Table 3). The 2 diagnosed patients in the imaging group did not demonstrate optic disc edema on funduscopic imaging. An additional 52 patients (4.8%) had symptoms of PTC (headaches and other symptoms) as indicated by a co-morbidity score of 1 or 2 (Table 3). Of the 45 patients with a score of Table 2 Patient demographics All patients Imaging group Nonimaging group Patients (n) Women 863 (79.6%) 414 (77.8%) 449 (81.3%) Age (yr) Weight (kg) BMI (kg/m 2 ) BMI body mass index P value not significant for all comparisons.

4 80 I. N. Hamdallah et al. / Surgery for Obesity and Related Diseases 9 (2013) Table 3 Co-morbidity scores Co-morbidity Score Imaging No imaging Diabetes mellitus Hypertension Dyslipidemia Sleep apnea syndrome Polycystic ovarian syndrome Pseudotumor cerebri Data in parentheses are percentages. Total (62.40) 339 (61.40) 671 (61.90) 1 56 (10.50) 64 (11.60) 120 (11.10) 2 84 (15.80) 90 (16.30) 174 (16.10) 3 24 (4.50) 12 (2.20) 36 (3.30) 4 34 (6.40) 40 (7.20) 74 (6.80) 5 2 (.40) 7 (1.30) 9 (.80) (43.60) 253 (45.80) 485 (44.70) 1 22 (4.10) 23 (4.20) 45 (4.20) 2 49 (9.20) 42 (7.60) 92 (8.50) (21.10) 115 (20.80) 226 (20.80) (21.80) 115 (20.80) 231 (21.30) 5 1 (.20) 4 (.70) 5 (.50) (50.40) 308 (55.80) 577 (53.20) 1 27 (5.10) 36 (6.50) 63 (5.80) 2 73 (13.70) 61 (11.10) 133 (12.30) (28.20) 126 (22.80) 276 (25.50) 4 11 (2.10) 19 (3.40) 30 (2.80) 5 3 (.60) 2 (.40) 5 (.50) (59.20) 326 (59.10) 641 (59.10) 1 66 (12.40) 67 (12.10) 133 (12.30) 2 16 (3.00) 16 (2.90) 32 (3.00) (25.00) 141 (25.50) 275 (25.40) 4 1 (.20) 1 (.20) 2 (.20) 5 1 (.20) 1 (.20) 1 (.10) (95.10) 513 (92.90) 1019 (94.00) 1 12 (2.20) 20 (3.80) 32 (3.00) 2 8 (1.50) 9 (1.60) 17 (1.60) 3 1 (.20) 7 (1.30) 8 (.70) 4 2 (.40) 1 (.20) 3 (.30) 5 3 (.60) 2 (.40) 5 (.50) (94.90) 522 (94.60) 1028 (94.80) 1 22 (4.10) 24 (4.30) 45 (4.20) 2 3 (.60) 4 (.70) 7 (.60) 3 0 (0) 0 (0) 0 (0) 4 2 (.40) 0 (0) 2 (.20) 5 0 (0) 2 (.40) 2 (.20) 1 for PTC, 22 had funduscopic imaging. Of these 22 patients, 4 (18%) had fundus images concerning for optic disc edema. Of the 4 patients, 2 were found to have optic disc edema on neuro-ophthalmologic evaluation. One patient underwent lumbar puncture, and 1 patient declined. Only 3 of the 7 patients with a score of 2 underwent funduscopic imaging, and none of these patients had optic disc edema. A total of 28 patients in the nonimaging group (24 with score of 1 and 4 with score of 2) had symptoms but did not undergo imaging (Table 3). Thus, our overall patient population of 1084 patients included 7 patients with PTC (3 detected by imaging and 4 previously diagnosed) for a prevalence of.65% and a calculated annual incidence of 323/100,000 (Table 4). Similarly, the imaging group of 532 patients included 5 patients with PTC (3 detected during the present study and 2 previously diagnosed) for a prevalence of.9% and a calculated annual incidence of 470/100,000 (Table 4). The patient who declined lumbar puncture was not included as a positive case in these calculations, because the diagnosis of PTC could not be confirmed. Furthermore, 2 of the patients who underwent lumbar puncture were asymptomatic (score 0), and their optic disc edema was detected on funduscopic screening. Discussion PTC is a well-recognized co-morbidity of obesity and responds well to bariatric surgery [2,11]. The increased incidence of PTC in obesity has been hypothesized to result from increased intra-abdominal pressure, which leads to increased intrathoracic pressure, which, in turn, leads to increased intracranial pressure [12,13]. Other hypotheses have implicated the systemic metabolic effects of obesity in the development of PTC, citing abnormalities in glucose metabolism, sex hormones, adipokines, glucocorticoids, lipids, and free fatty acids as potential mediators in this disease [2,14]. The disruption of cerebrospinal fluid (CSF) homeostasis has also been reported as a possible cause of PTC and includes decreased CSF absorption by arachnoid granulations, overproduction of CSF by the choroid plexus, and dysregulation of aquaporins (a family of regulatory membrane water channel proteins that participate in the secretion and reabsorption of CSF) [5]. An interesting recent report suggested that obesity might not cause PTC but might, in fact, be a consequence of PTC [15]. However, none of the current hypotheses regarding the etiology of PTC satisfactorily explains what has been observed clinically. The prevalence of PTC among Americans aged 15 years has been estimated to be.012% overall and increases to.028% among obese individuals (body mass index 30 kg/m 2 ) [16]. The incidence of PTC in the general population has been reported to be 1 2/100,000. This incidence can increase dramatically in specific patient populations, such as women aged years who weigh 20% over their ideal body weight (19.3/100,000) [8]. Some studies have suggested a significant correlation between increasing body weight and an increasing incidence of PTC [8,9]. Thus, it would be reasonable to hypothesize that PTC would be highly prevalent in bariatric surgical patients whose body mass, by definition, is significantly greater than overweight or class I obese patients. In the present 2-year study of patients with a body mass index of 35 kg/m 2, we found 7 patients with PTC, corresponding to an annual incidence of 323/100,000. Four of the

5 Prevalence of Pseudotumor Cerebri / Surgery for Obesity and Related Diseases 9 (2013) Table 4 Prevalence and incidence of PTC in all patients and patients imaged with funduscopic camera PTC detected PTC diagnosed Total (all patients with PTC) By imaging Before evaluation Patients (n) Prevalence (%) Annual incidence All patients (n 1084) /100,000 Imaging group (n 532) /100,000 PTC, pseudotumor cerebri. patients with PTC had been previously diagnosed and were being treated. Two of the treated patients underwent funduscopic imaging, which showed normal optic discs, probably reflecting adequate treatment. All 4 patients were symptomatic at the previous diagnosis. Funduscopic examination identified 3 additional patients with PTC, 2 of whom were completely asymptomatic. The third patient reported headaches and tinnitus during the co-morbidity screening. The prevalence of PTC in the present study was.65% overall and.9% in the imaging group. Because only 2 patients in the imaging group were asymptomatic before the discovery by funduscopic examination, the prevalence of asymptomatic PTC was.4%. Other than obesity, the greater than expected prevalence of PTC in the present study could have been related to the overall metabolic health of bariatric surgery patients. Metabolic dysregulation has been correlated with the development of PTC and has even been postulated to have an etiologic role in the development of this condition [2]. As indicated in Table 3, metabolic disorders (i.e., DM, HTN, DYS, SAS, PCOS), which were sufficiently severe to require medical treatment, occurred with some frequency in our patient population. DM (score 2) was present in 27%, HTN (score 3) was present in 43%, DYS (score 3) was present in 14%, SAS (score 2) was present in 29%, and PCOS (score 2) was present in 3%. These data support our notion that the imaging group had a similar metabolic predisposition to PTC as the nonimaging group and that we did not select patients with increased metabolic dysregulation for imaging. Furthermore, the metabolic co-morbidity profiles confirmed that patients evaluated by the bariatric surgeon express the significant risk factors that have been hypothesized to result in PTC. Despite the high incidence and prevalence in this patient population, PTC remains an uncommon disease, certainly much less common than other co-morbidities of obesity (Table 3). Clinically, obese patients should be screened for the signs and symptoms (e.g., headaches, dizziness, tinnitus) of PTC. Symptomatic patients should be evaluated further for the presence of papilledema and, if present, should be treated to prevent the visual sequelae of PTC. Although papilledema is the hallmark finding for PTC [2], an elevated intracranial pressure, in the absence of papilledema, has been reported but is rare [17]. Although identifying and treating symptomatic patients can be performed effectively, diagnosing asymptomatic patients poses a clinical challenge. Visual disturbances can be the initial presentation of PTC in some patients [18]. With the potential for such significant pathologic consequences, it would be important to identify the risk factors for PTC and screen patients deemed to be at high risk. Although it might be compelling to suggest a routine screening modality for all patients, our study found only marginal utility in the routine examination of asymptomatic patients. The results of our study also showed that major ocular abnormalities were infrequent in this population of bariatric surgical candidates, despite the frequent presence of significant medical co-morbidities (Table 3). Although PTC is an infrequent condition overall, it responds well to bariatric surgery. Also, although the diagnosis of PTC might not be critical in patients who are already scheduled to undergo bariatric surgery, many obese patients have undiagnosed PTC, which then is treated as regular headaches. These patients would benefit from bariatric surgery to prevent visual loss. Thus, it is important for physicians who treat obese individuals to consider this diagnosis in patients with persistent headaches and refer patients with PTC for the effective treatment of bariatric surgery. The present study was limited in that not all symptomatic patients were screened with funduscopic imaging. However, participation in the imaging group was entirely voluntary. Furthermore, 1 patient with optic disc edema declined lumbar puncture and, thus, could not be definitively evaluated for PTC. We believe that our overall study population of 1084 patients and the imaging group of 532 patients provided enough power to meaningfully address the hypothesis of this investigation. Conclusion PTC is an uncommon disease but is more prevalent among bariatric surgery candidates than would have been predicted from the current data. These patients should be screened for the signs and symptoms of PTC. No current effective method is available to identify PTC in asymptomatic patients, and we found routine funduscopic imaging to be of limited value. Asymptomatic PTC can be associated with visual impairment. These facts warrant future investigation into the utility of, and optimal method for, screening morbidly obese patients for this co-morbidity. Once the diagnosis of PTC has been established in the

6 82 I. N. Hamdallah et al. / Surgery for Obesity and Related Diseases 9 (2013) setting of morbid obesity, serious consideration should be given to bariatric surgery, which can effectively treat this condition. Disclosures The authors have no commercial associations that might be a conflict of interest in relation to this article. References [1] Ali MR, Fuller WD, Choi MP, Wolfe BM. Bariatric surgical outcomes. Surg Clin North Am 2005;85:835 52, vii. [2] Randhawa S, Van Stavern GP. Idiopathic intracranial hypertension (pseudotumor cerebri). Curr Opin Ophthalmol 2008;19: [3] Ball AK, Clarke CE. Idiopathic intracranial hypertension. Lancet Neurol 2006;5: [4] Dandy WE. Intracranial pressure without brain tumor: diagnosis and treatment. Ann Surg 1937;106: [5] Friedman DI, Jacobson DM. Diagnostic criteria for idiopathic intracranial hypertension. Neurology 2002;59: [6] Skau M, Brennum J, Gjerris F, Jensen R. What is new about idiopathic intracranial hypertension? An updated review of mechanism and treatment. Cephalalgia 2006;26: [7] Walker RW. Idiopathic intracranial hypertension: any light on the mechanism of the raised pressure? J Neurol Neurosurg, Psychiatry 2001;71:1 5. [8] Durcan FJ, Corbett JJ, Wall M. The incidence of pseudotumor cerebri: population studies in Iowa and Louisiana. Arch Neurol 1988;45: [9] Radhakrishnan K, Ahlskog JE, Garrity JA, Kurland LT. Idiopathic intracranial hypertension. Mayo Clin Proc 1994;69: [10] Ali MR, Maguire MB, Wolfe BM. Assessment of obesity-related comorbidities: a novel scheme for evaluating bariatric surgical patients. J Am Coll Surg 2006;202:70 7. [11] Sugerman HJ, Felton WL III, Sismanis A, Kellum JM, DeMaria EJ, Sugerman EL. Gastric surgery for pseudotumor cerebri associated with severe obesity. Ann Surg 1999;229: [12] Sugerman HJ, DeMaria EJ, Felton WL III, Nakatsuka M, Sismanis A. Increased intra-abdominal pressure and cardiac filling pressures in obesity-associated pseudotumor cerebri. Neurology 1997;49: [13] Sugerman HJ, Felton IW III, Sismanis A, et al. Continuous negative abdominal pressure device to treat pseudotumor cerebri. Int J Obes Relat Metab Disord 2001;25: [14] Ooi LY, Walker BR, Bodkin PA, Whittle IR. Idiopathic intracranial hypertension: can studies of obesity provide the key to understanding pathogenesis? Br J Neurosurg 2008;22: [15] Hannerz J, Ericson K. The relationship between idiopathic intracranial hypertension and obesity. Headache 2009;49: [16] Friesner D, Rosenman R, Lobb BM, Tanne E. Idiopathic intracranial hypertension in the USA: the role of obesity in establishing prevalence and healthcare costs. Obes Rev 2011;12:e [17] Digre KB, Nakamoto BK, Warner JE, Langeberg WJ, Baggaley SK, Katz BJ. A comparison of idiopathic intracranial hypertension with and without papilledema. Headache 2009;49: [18] Galvin JA, Van Stavern GP. Clinical characterization of idiopathic intracranial hypertension at the Detroit Medical Center. J Neurol Sci 2004;223: