Abstract

Background: Early treatment of cerebrospinal fluid (CSF) infections, including meningitis and/or ventriculitis (MV) is critical to minimize morbidity/mortality. Infections/MV are typically attributed to; recurrent postoperative lumbar CSF fistulas with drainage through the skin (12.2-33.3%), lumbar drains, and/or various intracranial devices (i.e. external ventricular drains, intracranial pressure monitors).

Methods: Lumbar MR examinations best document recurrent postoperative dural fistulas with subcutaneous extension leading to leaking wounds; the longer these leaks persist, the greater the risk of CSF infection and MV. Classical cranial MR findings of MV due to prior lumbar surgery, lumbar drains or multiple intracranial devices include; ventricular debris, ependymal enhancement, hydrocephalus, extra-axial fluid collections, infarcts (arteritis/ventriculitis), abscesses, and granulomas.

Results: Surgery for recurrent postoperative lumbar CSF leaks typically warrant wound reexploration with direct sutured-dural repairs, use of muscle patch grafts (avoid fat – it resorbs), fibrin sealants/fibrin glues (FS/FG), lumbar drains, lumboperitoneal and/or pseudomeningocele-peritoneal shunts. For patients who additionally develop meningitis/ventriculitis, one should consider adding intraventricular (IVT) or lumbar intrathecal (IT) antibiotic therapy to routine intravenous antibiotics. Notably, all efforts should be made to avoid the high mortality rates associated with VM (i.e., 13% to 60%).

Conclusion: Recurrent postoperative lumbar CSF leaks (i.e., especially after CSF breaches the skin), lumbar drains, and/or intracranial devices/implants may cause infections including meningitis and ventriculitis (MV). It is critical to recognize and treat these infections/MV early to avoid high morbidity and mortality rates.

Keywords: Cerebrospinal Fluid (CSF) Leaks, Dural Repair, Early Diagnosis, Early Surgery, External Fistulas, External Ventricular Drains, Fibrin Sealant/Fibrin Glues, Infection, Intracranial Pressure Monitors, Lumbar Drain, Lumbar Surgery, Meningitis, Magnetic Resonance Imaging (MR), Muscle Patch Graft, Traumatic Dural Tears (DT), Ventriculitis

INTRODUCTION

Intraoperative cerebrospinal fluid leaks occur in 2-20% of primary lumbar operations, but this frequency markedly increases to 13.3% to 33.3% for recurrent lumbar/secondary surgery [ Table 1 ].[ 1 , 4 , 15 ] Early diagnosis with MR and early surgery for postoperative recurrent fistulas (i.e. before CSF breaches the skin) are critical to reduce the risk of infection, including meningitis, and/or ventriculitis (VM) which can prove fatal.[ 4 , 5 , 8 , 14 ] Surgical options include; reexploration with direct repair of dural tears (DT), the addition of muscle patch grafts (avoid fat-it resorbs), the application of fibrin sealants/fibrin glues (FS/FG), the placement of external lumbar drains (LD), and/or shunts (i.e. lumboperitoneal and/or pseudomeningocele-peritoneal shunts).[ 3 , 4 , 6 , 15 ] For meningitis/ventriculitis, routine systemic intravenous antibiotic therapy should likely be supplemented with timely intraventricular (IVT) or lumbar intrathecal (IT) antibiotics in an attempt to reduce the VM mortality rates ranging from 13%-60%.[ 2 , 7 , 9 , 10 - 14 ]

Table 1:

Infection, Meningitis, and/or Ventriculitis Due to Recurrent CSF Leaks After Lumbar Surgery, Lumbar Drains, or Cranial Implants/Procedures (i.e. EVD, ICP Monitors, Other).

Incidence of Primary and Recurrent CSF Leaks with Lumbar Surgery

The risks of intraoperative cerebrospinal fluid leaks occurring during primary lumbar surgery ranges from 2-20%; notably, the rates for recurrent postoperative/secondary leaks is even higher ranging from 13.3% to 33.3% [ Table 1 ].[ 1 , 4 , 15 ] In Woodroffe et al. (2018) iatrogenic primary lumbar surgical CSF leaks occurred in 3.1% of patients (i.e., 124 of 3965 cases); 73.4% of 64 CSF leaks were effectively primarily repaired, while 87.8% of 49 patients were successfully managed with primary closure and the addition of lumbar drains.[ 15 ] They also observed that delayed reexplorations resulted in longer lengths of stay, higher readmission/reoperation/infection rates, and more overall adverse events. For Asplater et al. (2021) 11.8% of 187 patients undergoing minimally invasive lumbar surgery (MIS) had primary iatrogenic acute dural tears; however, in this series, as no CSF leaks breached the skin, there were no infections.[ 1 ] Interestingly, their incidence of primary MIS surgical CSF leaks was comparable to those for open procedures. In 2024, Epstein’s review showed a 2.6-8% incidence of primary lumbar CSF leaks, while postoperative recurrent/secondary fistulas occurred in a higher 0.83%-14.3% of patients.[ 4 ]

MR Diagnosis of Recurrent Postoperative/Secondary CSF Leaks Resulting in Infections Including Meningitis and Ventriculitis (MV)

MR studies with/without contrast are superior to CT examinations in identifying postoperative lumbar wound infections, including documenting meningitis and/or ventriculitis (MV) [ Table 1 ].[ 5 . 8 , 13 ] Kahamalla et al. (2000) described the efficacy of MR studies with/without contrast as capable of demonstrating fluid contiguous with the dura, and extending through the paraspinal soft tissues into the subcutaneous compartment.[ 8 ] They emphasized how critical it was to diagnose meningitis early as it could devolve into; hydrocephalus (i.e. noting that this could develop/be exacerbated particularly once external drainage was stopped (lumbar wounds repaired) and/or devices were removed), extra-axial masses, infarcts (arteritis, ventriculitis), abscesses, and granulomas, resulting in significant morbidity and mortality. In 2001, Fukui et al. (2001) emphasized the classical MR findings for pyogenic ventriculitis in 17 patients; 16 had ventricular debris, 13 exhibited periventricular irregularities, 13 had hydrocephalus, while other findings included; ependymal enhancement, meningitis, and/or abscess formation.[ 5 ] Further, Luque-Paz et al. (2021) treated 98 patients with ventriculitis including the 13.3% attributed to prior neurosurgical procedures; here, MR showed intraventricular pus in 82.7%, ependymal enhancement in 71.4%, and intraventricular loculations in 15.3% of patients.[ 13 ]

Recommendations for Treating Primary or Recurrent Postoperative Lumbar CSF Leaks

Two lumbar studies cited multiple surgical options for repairing primary, delayed, and/or recurrent postoperative CSF leaks [ Table 1 ].[ 3 , 4 ] Deen et al, (2003) treated 4 postoperative pseudomeningoceles using video-laparoscopy to place both (1) lumboperitoneal, and (2) pseudomeningocele-peritoneal shunts.[ 3 ] They claimed these combined procedures resulted in fewer reoperations/procedures (i.e. direct repairs, extended drainage, use of blood patches), shortened lenth of stay, results in more accurate shunt placement, and reduced operative times. For repairs of recurrent/secondary CSF leaks, Epstein (2024) advocated using an operating microscope, and combinations of non-resorbable 7-0 Gore-Tex interrupted sutures (running sutures more likely to fail/unfurl), muscle patch grafts (avoid fat-it resorbs), microfibrillar collagen, and fibrin sealant/fibrin glues (FS/FG).[ 4 ] In Epstein’s experience, only complex recurrent leaks warranted the placement of lumbar drains and/or lumboperitoneal shunts.

Treatment of Infections Including Use of IVT/IT Antibiotics for Treating Meningitis and Ventriculitis Attributed to Recurrent Postoperative Lumbar CSF Leaks, Lumbar Drains, and/or Intracranial Devices

Recurrent/secondary postoperative lumbar CSF leaks involving skin breaches and other (i.e. lumbar drains, intracranial drains/devices) causes of infections leading to MV require varied systemic and IVT/IT antibiotic regimens [ Table 1 ].[ 6 , 10 , 14 ] In Remes et al. series of 34 patients, mostly gram negative organisms caused wound infections/MV and required IVT/IT antibiotics.[ 14 ] For Lewin et al. (2019), of 105 ICU patients receiving IVT antibiotics for VM, 52.4% received Vancomycin, while 47.5% had IVT Aminoglycosides; 88.4% with both meningitis and ventriculitis achieved achieved CSF sterilization adding IVT with I VT Aminoglycosides vs. 9.5% with recurrent or persistent postoperative infections.[ 10 ] For the 232 out of 437 neurosurgical patients who developed drain-related meningitis/ventriculitis due to the placement of EVD, LD, and ICP monitors in Hussein et al. (2019) series, the main offending organism was Acinetobacter Baumannii.[ 6 ]

Mortality Rates for Meningitis/Ventriculitis Attributed to Postoperative Lumbar Recurrent CSF Leaks, Lumbar Drains, or Intracranial Devices/Implants

Varied high mortality rates (i.e.11.8% to 60%) were cited for patients with MV attributed to recurrent postoperative/secondary lumbar CSF leaks, lumbar drains, or intracranial devices [ Table 1 ].[ 2 , 10 , 12 - 14 ] For the 34 patients with MV treated with IVT/IT in Remes et al. series (2013), the mortality rate was 11.8% (i.e., however, the authors noted some of these some patients died from other causes).[ 14 ] There was a 18.1% mortality rate in Lewin et al. (2019) series of 105 MV patients undergoing IVT antibiotic treatment.[ 10 ] When Luque-Paz et al. (2021) treated 98 patients with ventriculitis, 13.3% of whom had previously undergone neurosurgical procedures, the overall mortality rate was 30.6%.[ 13 ] They found the in-hosptial 1 year mortality rate was 38.8%, and was associated with the following major risk factors; age over 65, Glasgow Coma score of > 13, Status Epilepticus, Hydrocephalus (i.e. developing concurrent with active lumbar CSF leaks/external drainage or in a delayed fashion once lumbar fistulas were repaired and/or after drains were removed), and positive CSF Cultures. Over 3 years, Coelho et al. (2021) found a 13% mortality rate among 218 neurosurgical patients with MV due to various intracranial devices.[ 2 ] Ippolito et al. (2022) recognized that MV infections could be attributed to neurosurgical CSF shunts, drains, implantable devices, and/or trauma, resulting in up to 60% mortality rates.[ 7 ] Lilley et al. (2022) observed a 14% mortality rate from ventriculitis that occurred in 99 patients treated with 105.90% ventricular devices (i.e. CSF shunts, external ventricular drains (EVD), and others).[ 12 ]

Treat Infections/MV Attributed to Recurrent Lumbar CSF Leaks, or Lumbar/Intracranial Implants with Intraventricular (IVT) or Lumbar Intrathecal (IT) Antibiotics

Several studies reviewed the diagnosis and treatments of CSF infections/MV attributed to recurrent postoperative lumbar CSF fistulas, or the placement of Lumbar Drains (LD), Intracranial External Ventricular Drains (EVD), or Intracranial Pressure Monitors (ICP) [ Table 1 ].[ 2 , 6 , 7 , 9 , 11 - 13 ] Remes et al. (2013) 34 patients with MV (i.e. 30 meningitis alone; 4 meningitis/ventriculitis) received IVT or IT (i.e. using a lumbar drain) antibiotics; CSF was sterilized within an average of 2.9 days; the CSF cultures were negative within 24 h with IVT/IT in 17 (50%) of patients, and in up to 48 hours for another 6 (18%) patients.[ 14 ] Outcomes using the Modified Rankin Scale showed; improvement in 17 patients (50%), no change 10 patients (29%), 1 poor result (3%) and 6 deaths. They concluded IVT/IT resulted in rapid sterilization of CSF for patients with MV. In Hussein et al. (2019), 232 out of 437 neurosurgical patients developed drain-related meningitis/ventriculitis attributed to EVD, LD, and ICP monitors; the infection rate was 13.7 per 1000 drain days, and the predominant organism was Acinetobacter Baumannii.[ 6 ] Risk factors for MV included; diabetes and external CSF leaks, with the most prominent factor including the frequency of drain opening, and longer duration of drain use; ultimate recommendations included shorter duration of drainage with fewer openings. Luque-Paz et al. in 2021 had 98 patients with ventriculitis variously attributed to; 13.3% neurosurgical procedures; 20.6% after treatment of brain abscess, 27.6% with meningitis, and 7l% with intraventricular catheters.[ 13 ] They identified the following organisms; Streptococcus 44.9%, Gram Negatives 27.6%, and Staphylococcus 15.3%). Ultimately, 61.8% ended up with residual neurological deficits that included; 9 with gait disturbance, 7 with paresis, 5 with seizures, and 6 with behavioral and/or cognitive changes. Coelho et al. (2021) diagnosed 218 VM infections over 3 years in patients with various intracranial devices exhibiting persistent fevers, and high CSF cells counts.[ 2 ] Karyouniaris et al, (2022) diagnosed VM in post neurosurgery patients, and managed them with IVT (i.e. they had imlanted devices including catheters and drains); most infections were due to Gram Negative bacteria (i.e., skin flora, or other nosocomial pathogens).[ 9 ] They did observe the following risk factors associated with using IVT; seizures, aseptic meningitis, and biofilm on implanted catheters prolonging or increasing the risks of persistent infection. Ippolito et al. (2022) recognized that VM could develop due to various neurosurgical devices (i.e. CSF shunts, drains, implantable devices, and after trauma); notably, once external drainage of CSF was occluded (i.e. lumbar wounds repaired or lumbar drains/EVDs removed) patients who had developed underlying hydrocephalus often subsequently required placement of ventriculoperitoneal or lumboperitoneal shunts. Organisms identified typically included. Gram Positives, and only rarely, were multiple resistant organisms encountered.[ 7 ] Lilley et al. (2022) analyzed 99 patients with 105.90% ventricular devices (i.e., due to CSF shunts, EVD and others), and advocated, IVT with device removal.[ 12 ] The most common organisms were Coagulase Negative Staph, Gram Negative Bacteria, or Gram Positive Bacteria. Typically, the wrong antibiotics were being given (Meropenem and Linezolid). Li et al. (2024) recommended early IVT antibiotic installation to reduce inhospital mortality due to MV; IVT was additionally given to 102 of 142 patients with MV receiving systemic antibiotics.[ 11 ] This resulted in higher rates and shorter times to sterilization of CSF, better outcomes, and lower mortality rates.

CONCLUSION

CSF infections resulting in meningitis/ventriculitis may be attributed to recurrent postoperative lumbar CSF leaks (i.e., ideally prior to CSF breaching the skin), lumbar drains, and/or cranial implants/devices. Early diagnosis and treatment of these infections/MV will likely reduce their accompanying high morbidity and mortality rates.

Ethical approval:

The Institutional Review Board approval is not required.

Declaration of patient consent:

Patient’s consent was not required as there are no patients in this study.

Financial support and sponsorship:

Nil.

Conflicts of interest:

There are no conflicts of interest.

Use of artificial intelligence (AI)-assisted technology for manuscript preparation:

The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.

Disclaimer

The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Journal or its management. The information contained in this article should not be considered to be medical advice; patients should consult their own physicians for advice as to their specific medical needs.

References

1. Aspalter S, Senker W, Randl C, Aichholzer M, AufschnaiterHiebbock K, Leitner C. Accidental dural tears in minimally invasive spinal surgery for degenerative lumbar spine disease. Front Surg. 2021. 8: 708243

2. Coelho E, Costa L, Martins J, Costa M, Oliveira JE, MaiaGoncalves A. Healthcare-associated ventriculitis and meningitis: A retrospective analysis. Cureus. 2021. 13: e19069

3. Deen HG, Pettit PD, Sevin BU, Wharen RE, Reimer R. Lumbar peritoneal shunting with video-laparoscopic assistance: A useful technique for the management of refractory postoperative lumbar CSF leaks. Surg Neurol. 2003. 59: 473-7 discussion 477-8

4. Epstein NE, Agulnick MA. Perspective: Timely diagnosis and repair of intraoperative thoracic/lumbar cerebrospinal fluid (CSF) leaks. Surg Neurol Int. 2024. 15: 255

5. Fukui MB, Williams RL, Mudigonda S. CT and MR imaging features of pyogenic ventriculitis. AJNR Am J Neuroradiol. 2001. 22: 1510-6

6. Hussein K, Rabino G, Feder O, Eghbaryeh H, Zayyad H, Syiri G. Risk factors for meningitis in neurosurgical patients with cerebrospinal fluid drains: Prospective observational cohort study. Acta Neurochir (Wien). 2019. 161: 517-24

7. Ippolito M, Giarrantano A, Cortegiani A. Healthcare-associated central nervous system infections. Curr Opin Anaesthesiol. 2022. 35: 549-54

8. Kanamalla US, Ibarra RA, Jinkins JR. Imaging of cranial meningitis and ventriculitis. Neuroimaging Clin N Am. 2000. 10: 309-31

9. Karyouniaris M, Brotis A, Tsiakos K, Palli E, Koulenti D. Current perspectives on the diagnosis and management of healthcare-associated ventriculitis and meningitis. Infect Drug Resist. 2022. 15: 697-721

10. Lewin JJ, Cook AM, Gonzales C, Merola D, Neyens R, Peppard WJ. Current practices of intraventricular antibiotic therapy in the treatment of meningitis and ventriculitis: Results from a multicenter retrospective cohort study. Neurocrit Care. 2019. 30: 609-16

11. Li Z, Yang W, Ye X, Yuan Q, Zhao J, Du Z. Early intraventricular antibiotic therapy improved in-hospital-mortality in neurocritical patients with multidrug-resistant bacterial nosocomial meningitis and ventriculitis. Neurocrit Care. 2024. 40: 612-20

12. Lilley D, Munthali P. Analysis of the management of ventriculitis cases at a UK neurosurgery centre. Infect Prev Pract. 2022. 4: 100240

13. Luque-Paz D, Revest M, Eugene F, Boukthir S, Dejoies L, Tattevin P. Ventriculitis: A severe complication of central nervous system infections. Open Forum Infect Dis. 2021. 8: ofab216

14. Remes F, Tomas R, Jindrak V, Vanis V, Setlik M. Intraventricular and lumbar intrathecal administration of antibiotics in postneurosurgical patients with meningitis and/or ventriculitis in a serious clinical state. J Neurosurg. 2013. 119: 1596-602

15. Woodroffe RW, Nourski KV, Helland LC, Walsh B, Noeller J, Kerezoudis P. Management of iatrogenic spinal cerebrospinal fluid leaks: A cohort of 124 patients. Clin Neurol Neurosurg. 2018. 170: 61-6