Elsevier

World Neurosurgery

Volume 103, July 2017, Pages 275-282
World Neurosurgery

Original Article
Postoperative Neurosurgical Infection Rates After Shared-Resource Intraoperative Magnetic Resonance Imaging: A Single-Center Experience with 195 Cases

https://doi.org/10.1016/j.wneu.2017.03.093Get rights and content

Objectives

To determine the rate of surgical-site infections (SSI) in neurosurgical procedures involving a shared-resource intraoperative magnetic resonance imaging (ioMRI) scanner at a single institution derived from a prospective clinical quality management database.

Methods

All consecutive neurosurgical procedures that were performed with a high-field, 2-room ioMRI between April 2013 and June 2016 were included (N = 195; 109 craniotomies and 86 endoscopic transsphenoidal procedures). The incidence of SSIs within 3 months after surgery was assessed for both operative groups (craniotomies vs. transsphenoidal approach).

Results

Of the 109 craniotomies, 6 patients developed an SSI (5.5%, 95% confidence interval [CI] 1.2–9.8%), including 1 superficial SSI, 2 cases of bone flap osteitis, 1 intracranial abscess, and 2 cases of meningitis/ventriculitis. Wound revision surgery due to infection was necessary in 4 patients (4%). Of the 86 transsphenoidal skull base surgeries, 6 patients (7.0%, 95% CI 1.5–12.4%) developed an infection, including 2 non-central nervous system intranasal SSIs (3%) and 4 cases of meningitis (5%). Logistic regression analysis revealed that the likelihood of infection significantly decreased with the number of operations in the new operational setting (odds ratio 0.982, 95% CI 0.969–0.995, P = 0.008).

Conclusions

The use of a shared-resource ioMRI in neurosurgery did not demonstrate increased rates of infection compared with the current available literature. The likelihood of infection decreased with the accumulating number of operations, underlining the importance of surgical staff training after the introduction of a shared-resource ioMRI.

Introduction

During the past decade, intraoperative magnetic resonance imaging (ioMRI) increasingly has been implemented for numerous neurosurgical procedures.1, 2, 3, 4, 5, 6, 7 Although the use of high-field ioMRI has greatly influenced the management of neuro-oncological patients by increasing the extent of tumor resection and therefore survival time,8 the acquisition and maintenance costs of high-field scanners are considerably high, imposing a challenge for many neurosurgical clinics.9 As a cost-effective alternative to the 1-room ioMRI, the 2-room ioMRI suite comprises a high-field scanner that is spatially separated from the operating theater by a double-layered sliding door (Figure 1). This setting allows for continuous nonsurgical outpatient magnetic resonance imaging (MRI) examinations outside the ioMRI time window, thereby sharing resources and greatly improving the scanner's cost-effectiveness.10

Although procedures in 1-room ioMRI suites have infection rates as low as conventional operations,11, 12 there are concerns whether such a shared-resource design imposes a greater risk for infections or anesthesia-related complications.13 For example, patient transportation into the MRI room imposes a challenge to anesthesia, and it is still unclear whether exposure to a nonsterile environment such as the ioMRI, the repetitive draping and undraping of the operation site, as well as the longer procedure time of approximately 1 hour increase the risk of surgical-site infections (SSIs).13 Although several studies find low infection rates in shared-resource designs,9, 14, 15, 16 their validity is restricted because they only investigated a small number of patients and rarely included transsphenoidal procedures.

This study is part of an ongoing quality assessment of neurosurgical complication rates at our clinic.17 Specifically, in this study we prospectively documented SSIs after 195 consecutive procedures involving a 2-room ioMRI. The rates of infections and infection-related reoperations of 2 operative groups (craniotomies and transsphenoidal approaches) were then compared with rates from previous studies without ioMRIs and single-room scanners to assess whether the use of this 2-room setting imposes a greater infection risk for patients. Furthermore, logistic regression analysis was used to determine whether the likelihood of infection changes with the accumulative number of operations performed in this new setting.

Section snippets

Patients

We prospectively documented all 195 consecutive neurosurgical interventions with our 2-room ioMRI that occurred between April 2013 and June 2016. Ethical board approval was given (KEK-ZH 2012-0244). Operations were performed by 11 surgeons at the University Hospital Zurich, led by one of the two senior authors. The use of the ioMRI was decided on medical indication and the surgeon's preference. The surgeries were grouped by approach (109 craniotomies on 102 patients and 86 transsphenoidal skull

Patient Cohorts and Operative Statistics

Patient characteristics are listed in Table 1. The average age of the craniotomy group was considerably lower than in the transsphenoidal group (40 [21, 57] vs. 46 [33, 63.75] years median age and quartiles) because of the high proportion of pediatric patients in this cohort (22%). The majority of the procedures were first-time operations (83% of the transsphenoidal procedures, 58% of the craniotomies). Craniotomies were performed for revascularization (10%) and resection of tumors (glioma in

Incidence of SSIs in Neurosurgical Operations with Shared-Resource ioMRI Is within Normal Range

In this study, we have evaluated the incidence of postoperative infections in neurosurgical procedures that involved a shared-resource ioMRI. During the study period, we performed 109 craniotomies and 86 transsphenoidal skull-base surgeries with involvement of a shared-resource ioMRI and observed 6 cases of SSI (5.5%) among patients who underwent craniotomy and 6 cases of postoperative infection among patients who underwent transsphenoidal surgery (7.0%), although only 4 of the infections in

Conclusions

Despite transportation into a nonsterile environment and prolonged operation time, the use of a 2-room ioMRI in neurosurgical procedures did not demonstrate increased infection rates compared with other studies. However, the incidence of postoperative infections in both study groups was initially greater than targeted. The likelihood of infection significantly decreased with the number of operations, underlining the importance of surgical staff training after the introduction of the novel

References (37)

  • A. Borg et al.

    Endoscopic endonasal anterior skull base surgery: a systematic review of complications over the past 65 years

    World Neurosurg

    (2016)
  • J. Burkhardt et al.

    Intraoperative low-field MR-guided frameless stereotactic biopsy for intracerebral lesions

    Acta Neurochir (Wien)

    (2013)
  • F. Cornaz et al.

    Compatibility of intraoperative 3T MR imaging and intraoperative neurophysiological monitoring

    J Neurol Surg A Cent Eur Neurosurg

    (2014)
  • J. Fierstra et al.

    Blood oxygen-level dependent functional assessment of cerebrovascular reactivity: feasibility for intraoperative 3 Tesla MRI

    Magn Reson Med

    (2017)
  • M. Neidert et al.

    The influence of intraoperative resection control modalities on survival following gross total resection of glioblastoma

    Neurosurg Rev

    (2016)
  • C. Serra et al.

    Pituitary surgery and volumetric assessment of extent of resection: a paradigm shift in the use of intraoperative magnetic resonance imaging

    Neurosurg Focus

    (2016)
  • R. Ramina et al.

    Optimizing costs of intraoperative magnetic resonance imaging. A series of 29 glioma cases

    Acta Neurochir (Wien)

    (2009)
  • M.N. Pamir

    3 T ioMRI: The Istanbul experience

  • Cited by (18)

    • The Relationship Between Colloid Transfusion During Surgical Decompression Hemicraniectomy Period and Postoperative Pneumonia or Long-Term Outcome After Space-Occupying Cerebral Infarction: A Retrospective Study

      2019, World Neurosurgery
      Citation Excerpt :

      Patient-specific data were collected as shown in our previous study.13 The analyzed variables included race; age; sex; height; weight; initial National Institutes of Health Stroke Scale (NIHSS) score after admission into operating room; interval from the onset of stroke to operation; atrial fibrillation; history of hypertension; type II diabetes mellitus; tobacco use; duration of surgery; and urine volume; 6% HES 130/0.4 transfusion; crystalloid transfusion (sodium acetate ringer); positive balance fluids administered before, during, and after (within the first day) surgery; systolic blood pressure on admission; hematocrit, PO2, glucose on admission, and during surgical treatment; postoperative complications (delayed intracranial hematomas, acute respiratory distress syndrome, pneumonia, sepsis, and AKI); and length of hospital stay, which were systematically confirmed by clinical criteria according to previous studies.13-15 One year after surgical treatment, the status of patients with space-occupying cerebral infarction was determined based on the modified Rankin Scale (mRS) according to the record of a personal interview by clinic visit or telephone.

    View all citing articles on Scopus

    Conflict of interest statement: The Department of Neurosurgery and the Department of Neuroradiology at the University Hospital Zurich are reference centers for ioMRI scanners of Siemens AG.

    View full text