Issue Navigator

Volume 15 No. 05
Earn CME
Accepted Papers

Case Reports

Iatrogenic Nontraumatic CPAP-Induced Pneumocephalus in a Patient With Meningitis

Antonios Charokopos, MD, MSc1; Mary Elizabeth Card, MD1; Richard Peter Manes, MD2; Albert Shaw, MD, PhD3
1Department of Internal Medicine, Yale-New Haven Hospital, Yale School of Medicine, New Haven, Connecticut; 2Department of Otolaryngology, Yale-New Haven Hospital, Yale School of Medicine, New Haven, Connecticut; 3Section of Infectious Diseases, Department of Internal Medicine, Yale-New Haven Hospital, Yale School of Medicine, New Haven, Connecticut


Pneumocephalus, or air within the cranium, can be caused by trauma, intracranial infections, or tumors, and can also occur as a complication of neurosurgery and lumbar puncture. Continuous positive airway pressure (CPAP) therapy can precipitate or worsen pneumocephalus in cases of known head trauma. However, nontraumatic pneumocephalus being caused by CPAP is a highly unexpected clinical event. We describe a case of a patient who presented with meningitis due to an atypical organism that usually resides in the oral cavity, and who developed nontraumatic pneumocephalus in the hospital due to CPAP therapy. The underlying cause, a cerebrospinal fluid leak, was likely the mediator for both pathologies. In the setting of the increasing prevalence of obstructive sleep apnea, physicians can benefit from being aware of this atypical presentation of meningitis and atypical complication of CPAP therapy.


Charokopos A, Card ME, Manes RP, Shaw A. Iatrogenic nontraumatic CPAP-induced pneumocephalus in a patient with meningitis. J Clin Sleep Med. 2019;15(5):781–783.


With the increasing prevalence of obesity and increase in physician recognition, the diagnosis of obstructive sleep apnea (OSA) has become common among the general population. The cornerstone of OSA treatment, continuous positive airway pressure (CPAP), provides continuous pressure in order to stent the airway open and relieve any obstruction. CPAP is a therapy that is usually considered benign. Classically in the literature, positive airway pressure therapy or bag mask ventilation have in certain situations caused pneumocephalus when a prior physical force has disrupted the base of the skull, specifically in traumatic accidents13 or postoperatively after transsphenoidal surgery4 or even years after posttraumatic surgery.5 Nontraumatic pneumocephalus from CPAP is extremely rare, and has only been reported twice so far in the literature.6,7 In both cases, there was a presumed cerebrospinal fluid (CSF) leak, although no surgical intervention was pursued and meningitis failed to develop in neither patient as a cause or a consequence of the CSF leak. In this report, we discuss a case of nontraumatic pneumocephalus, in the context of active meningitis and subsequent CPAP use.


A 67-year-old woman with history of severe OSA (apnea-hypopnea index 84 events/h, oxygen saturation nadir 81%, 28% time with oxygen saturation below 90%), uncontrolled hyper-tension, diabetes, and a known incidental 3 mm left paraclinoid internal carotid artery aneurysm presented with lethargy, headache, photophobia, rhinorrhea, and chills. On presentation, she was noted to have a temperature of 103.6°F (39.8°C). A computed tomography (CT) scan of the head obtained on presentation was normal. Lumbar puncture was significant for 6,905 nucleated cells/μL (94% granulocytes), protein 185 mg/dL (normal < 50 mg/dL), and glucose 76 mg/dL (normal 40–70 mg/dL), and the patient was started on empiric meningitis-appropriate antimicrobial coverage. Because the patient was allergic to penicillin and cephalosporin (reporting severe kidney injury from both), the combination of vancomycin, chloramphenicol, and acyclovir was used. Her CSF cultures grew viridans group streptococci (sensitive to ceftriaxone and vancomycin, but with intermediate susceptibility to penicillin). Blood cultures were negative. The patient's antimicrobial drugs were subsequently narrowed to vancomycin.

The patient had been nonadherent with CPAP at home (adherence 7.8%), but because of her OSA she was empirically placed on CPAP 10 cmH2O while in the hospital. Sleep or pulmonary consultation was not requested. Her headache persisted even after many days of antibiotic treatment, so on hospital day 6 an MRI of the head was performed. Interestingly, unlike her normal CT scan on presentation, the MRI not only showed leptomeningeal enhancement consistent with her known meningitis, but also subarachnoid air. A subsequent CT scan of the head confirmed diffuse subarachnoid air. A sinus CT showed trace air within the left olfactory fossa and trace fluid involving the posterior olfactory recess, concerning for an underlying cribriform plate defect (Figure 1). The CPAP was discontinued. In the interim, the rhinorrhea developed in this patient. Her nasal fluid was tested and returned positive for beta-2 transferrin, confirming a CSF leak. The patient was discharged to complete a total course of 2 weeks of vancomycin, with plans for outpatient surgery for her skull base defect and CSF leak. Unfortunately, she was readmitted 4 days after discharge with worsening headaches and increased pneumocephalus. CT showed persistent subarachnoid air (Figure 2). She had finished her antibiotic course the day prior to readmission, and repeat lumbar puncture off antibiotics showed resolution of her CSF neutrophilic predominance (26 nucleated cells/μL, protein 42 mg/dL, glucose 75 mg/dL) and CSF cultures were negative; thus the mild pleocytosis was attributed to resolving inflammation. Vancomycin was continued until surgical repair via nasal endoscopy and stereotactic computer-assisted navigation. Using fluorescein injection into the CSF, the surgical procedure confirmed a left ethmoid CSF leak medial to the superior turbinate. The defect was repaired using a mucosal graft, and the patient had an uneventful postoperative recovery.

Computed tomography of the patient's sinuses is concerning for a cribriform plate defect (arrow).


Figure 1

Computed tomography of the patient's sinuses is concerning for a cribriform plate defect (arrow).

(more ...)

Subarachnoid air with pneumocephalus on sagittal view.


Figure 2

Subarachnoid air with pneumocephalus on sagittal view.

(more ...)


The source of the viridans streptococci, an unusual cause for community-acquired bacterial meningitis, was initially unclear. However, the development of pneumocephalus, a very rare complication of CPAP therapy without a preceding history of trauma, combined with CSF rhinorrhea led to the identification of a cribriform plate defect causing a CSF leak. There can be two hypotheses for these findings: (1) there was a spontaneous CSF leak that allowed for meningitis to occur and the pneumocephalus was iatrogenic from CPAP use, and (2) there was no CSF leak prior to the occurrence of meningitis and the application of CPAP against an inflamed meningocele caused a new CSF leak. We believe the first hypothesis is the most likely explanation, as viridans streptococci is an oral organism and rarely causes meningitis unless there is contiguous spread. Furthermore, there is not always a pressing need for CPAP therapy for OSA in inpatients with acute infection when they have been nonadherent in the outpatient setting; ordering inpatient CPAP for OSA should include a careful balancing of the benefits versus the potential perils associated with its use.

The patient's initial CT scan in the emergency room was normal, without any intracranial air. The pneumocephalus developed only after inpatient initiation of CPAP, suggesting this was iatrogenic. We suggest that a search for a potential CSF leak and temporary withholding of CPAP therapy be pursued in patients with meningitis due to oral atypical organisms.


Work for this study was performed at the Yale-New Haven Hospital, Yale School of Medicine. All authors have seen and approved the manuscript. The authors report no conflicts of interest.



continuous positive airway pressure


cerebrospinal fluid


computed tomography


obstructive sleep apnea



Klopfenstein CE, Forster A, Suter PM. Pneumocephalus. A complication of continuous positive airway pressure after trauma. Chest. 1980;78(4):656–657. [PubMed]


Nicholson B, Dhindsa H. Traumatic tension pneumocephalus after blunt head trauma and positive pressure ventilation. Prehosp Emerg Care. 2010;14(4):499–504. [PubMed]


van Heerden PV, Pinder M, Cameron PD. Pneumocephalus and resuscitation. Anaesth Intensive Care. 1999;27(2):223–224. [PubMed]


Kopelovich JC, de la Garza GO, Greenlee JD, Graham SM, Udeh CI, O'Brien EK. Pneumocephalus with BiPAP use after transsphenoidal surgery. J Clin Anesth. 2012;24(5):415–418. [PubMed Central][PubMed]


Jones RM, Coleman J, Davis EM. An unusual headache in a patient treated with CPAP. American Thoracic Society website. Published December 2017. Accessed April 15, 2019.


Jarjour NN, Wilson P. Pneumocephalus associated with nasal continuous positive airway pressure in a patient with sleep apnea syndrome. Chest. 1989;96(6):1425–1426. [PubMed]


Bernstein AL, Cassidy J, Duchynski R, Eisenberg SS. Atypical headache after prolonged treatment with nasal continuous positive airway pressure. Headache. 2005;45(5):609–611. [PubMed]