John C. Wellons III
John C. Wellons III and R. Shane Tubbs
Travis R. Ladner, Ashly C. Westrick, John C. Wellons III and Chevis N. Shannon
The purpose of this study was to design and validate a patient-reported health-related quality of life (HRQOL) instrument for pediatric Chiari Type I malformation (CM-I), the Chiari Health Index for Pediatrics (CHIP).
The CHIP has 45 items with 4 components making up 2 domain scores, physical (pain frequency, pain severity, nonpain symptoms) and psychosocial; physical and psychosocial scores are combined to create an overall HRQOL score. Increasing scores (0 to 1) represent increasing HRQOL. Fifty-five patients with CM-I (mean age 12 ± 4 years, 53% male) were enrolled and completed the CHIP and Health Utilities Index Mark 3 (HUI3). Twenty-five healthy controls (mean age 11.9 ± 4 years, 40% male) also completed the CHIP. CHIP scores were compared between these groups via the Mann-Whitney U-test. For CHIP discriminative function, subscore versus presence of CM-I was compared via receiver operating characteristic curve analysis. CHIP scores in the CM-I group were stratified by symptomatology (asymptomatic, headaches, and paresthesias) and compared via Kruskal-Wallis test with Mann-Whitney U-test with Bonferroni correction (p < 0.0167). CHIP was compared with HUI3 (Health Utilities Index Mark 3) via univariate and multivariate linear regression.
CHIP physical and psychosocial subscores were, respectively, 24% and 18% lower in CM-I patients than in controls (p < 0.001); the overall HRQOL score was 23% lower as well (p < 0.001). The area under the curve (AUC) for CHIP physical subscore versus presence of CM-I was 0.809. CHIP physical subscore varied significantly with symptomatology (p = 0.001) and HUI3 pain-related quality of life (R2 = 0.311, p < 0.001). The AUC for CHIP psychosocial subscore versus presence of CM-I was 0.754. CHIP psychosocial subscore varied significantly with HUI3 cognitive- (R2 = 0.324, p < 0.001) and emotion-related (R2 = 0.155, p = 0.003) quality of life. The AUC for CHIP HRQOL versus presence of CM-I was 0.820. Overall CHIP HRQOL score varied significantly with symptomatology (p = 0.001) and HUI3 multiattribute composite HRQOL score (R2 = 0.440, p < 0.001).
The CHIP is a patient-reported, CM-I-specific HRQOL instrument, with construct validity in assessing pain-, cognitive-, and emotion-related quality of life, as well as symptomatic features unique to CM-I. It holds promise as a discriminative HRQOL index in CM-I outcomes assessment.
Michael C. Dewan, Jaims Lim, Chevis N. Shannon and John C. Wellons III
Up to one-third of patients with a posterior fossa brain tumor (PFBT) will experience persistent hydrocephalus mandating permanent CSF diversion. The optimal hydrocephalus treatment modality is unknown; the authors sought to compare the durability between endoscopic third ventriculostomy (ETV) and ventriculoperitoneal shunt (VPS) therapy in the pediatric population.
The authors conducted a systematic review of articles indexed in PubMed between 1986 and 2016 describing ETV and/or VPS treatment success/failure and time-to-failure rate in patients < 19 years of age with hydrocephalus related to a PFBT. Additionally, the authors conducted a retrospective review of their institutional series of PFBT patients requiring CSF diversion. Patient data from the systematic review and from the institutional series were aggregated and a time-to-failure analysis was performed comparing ETV and VPS using the Kaplan-Meier method.
A total of 408 patients were included from 12 studies and the authors' institutional series: 284 who underwent ETV and 124 who underwent VPS placement. The analysis included uncontrolled studies with variable method and timing of CSF diversion and were subject to surgeon bias. No significant differences between cohorts were observed with regard to age, sex, tumor grade or histology, metastatic status, or extent of resection. The cumulative failure rate of ETV was 21%, whereas that of VPS surgery was 29% (p = 0.105). The median time to failure was earlier for ETV than for VPS surgery (0.82 [IQR 0.2–1.8] vs 4.7 months [IQR 0.3–5.7], p = 0.03). Initially the ETV survival curve dropped sharply and then stabilized around 2 months. The VPS curve fell gradually but eventually crossed below the ETV curve at 5.7 months. Overall, a significant survival advantage was not demonstrated for one procedure over the other (p = 0.21, log-rank). However, postoperative complications were higher following VPS (31%) than ETV (17%) (p = 0.012).
ETV failure occurred sooner than VPS failure, but long-term treatment durability may be higher for ETV. Complications occurred more commonly with VPS than with ETV. Limited clinical conclusions are drawn using this methodology; the optimal treatment for PFBT-related hydrocephalus warrants investigation through prospective studies.
R. Shane Tubbs, W. Jerry Oakes, John C. Wellons III and Paul A. Grabb
Nicholas Ettinger, Matthew Pearson, Fred S. Lamb and John C. Wellons III
In this report, the authors describe the case of a teenage boy who presented with hypertensive emergency, posterior reversible encephalopathy syndrome, and hydrocephalus due to fourth ventricle outlet obstruction. Posterior reversible encephalopathy syndrome is a well-characterized but uncommon syndrome in children that is generally triggered by severe hypertension. The unusual clinical picture of this patient, who had isolated cerebellar edema leading to obstructive hydrocephalus, has been rarely described in children.
R. Shane Tubbs, E. George Salter, John C. Wellons III, Jeffrey P. Blount and W. Jerry Oakes
Object. Reports describing surgical landmarks with which to identify the branches of the lumbar plexus found on the posterior abdominal wall are lacking in the English-language literature.
Methods. The authors examined 22 sides from six female and five male cadavers. Measurements were made between the branches of the lumbar plexus and various bone landmarks such as the midline vertebral bodies, supracristal plane—a horizontal line connecting the superior-most aspect of the left and right iliac crests approximating the L4–5 vertebrae—and anterior superior iliac spine (ASIS).
The mean distances from the midline and as they emerged through or lateral to the psoas major muscle to the subcostal, iliohypogastric, ilioinguinal, lateral femoral cutaneous, genitofemoral, and femoral nerves, were 5.5, 6, 6.5, 6, 4.5, and 4.5 cm, respectively. At a vertical line through the midpoint between the ASIS and the midline, the subcostal, iliohypogastric, and ilioinguinal nerves were superior to the supracristal plane at mean distances of 8, 4, and 5 cm, respectively. Inferior to the supracristal plane and in a vertical line through a midpoint between the ASIS and the midline, the lateral femoral and femoral nerves were found to have mean distances of 5 and 5.5 cm, respectively. The obturator nerve had a mean distance of 3 cm lateral to the midline. Additionally, the lateral femoral cutaneous nerve had a mean distance of 1.5 cm inferomedial to the ASIS.
Conclusions. A good working knowledge of the locations and anatomy of the nerves of the lumbar plexus on the posterior abdominal wall is necessary for the surgeon who operates in this region. The measurements provided herein will aid the surgeon who wishes to expose or avoid these nerves, thus precluding injury.
Robert P. Naftel, Gavin T. Reed, Abhaya V. Kulkarni and John C. Wellons III
Endoscopic third ventriculostomy (ETV) success is dependent on patient characteristics including age, origin of hydrocephalus, and history of shunt therapy. Using these factors, an Endoscopic Third Ventriculostomy Success Score (ETVSS) model was constructed to predict success of therapy. This study reports a single-institution experience with ETV and explores the ETVSS model validity.
A retrospective chart review identified 151 consecutive patients who underwent ETV at a pediatric hospital between August 1995 and December 2009. Of these 151, 136 patients had at least 6 months of clinical follow-up. Data concerning patient characteristics, operative characteristics, radiological findings, complications, and success of ETV were collected. The actual success rates were compared with those predicted by the ETVSS model.
The actual success rate of ETV at 6 months was 68.4% (93 of 136 patients), which compared well to the predicted ETVSS of 76.5% ± 12.5% (± SD). The C-statistic was 0.74 (95% CI 0.65–0.83), suggesting that the ability of the ETVSS to discriminate failures from successes was good. Secondary ETV was found to have a hazard ratio for failure of 4.2 (95% CI 2.4–7.2) compared with primary ETV (p < 0.001). The complication rate was 9.3% with no deaths. At the first radiological follow-up, the increased size of ventricles had a hazard ratio for failure of 3.0 (95% CI 1.5–6.0) compared with patients in whom ventricle size either remained stable or decreased (p = 0.002).
The ETVSS closely predicts the actual success of ETV, fitting the statistical model well. Shortcomings of the model were identified in overestimating success in patients with ETVSS ≤ 70, which may be attributable to the poor success of secondary ETVs in the authors' patient population.
R. Shane Tubbs, E. George Salter, John C. Wellons III, Jeffrey P. Blount and W. Jerry Oakes
There is a paucity of information in the neurosurgical literature regarding the surgical anatomy surrounding the posterior interosseous nerve (PIN). The goal of the current study was to provide easily recognizable superficial bone landmarks for identification of the PIN.
Thirty-four cadaveric upper extremities obtained from adults were subjected to dissection of the PINs, and measurements were made between this nerve and surrounding superficial bone landmarks.
In all specimens the main radial trunk was found to branch into its superficial branch and PIN at the level of the lateral epicondyle of the humerus. Proximally, the PIN was best identified following dissection between the brachioradialis and extensor carpi radialis longus and brevis muscles. At its exit site from the supinator muscle, the PIN was best identified after retraction between the extensor carpi radialis longus and brevis and extensor digitorum communis muscles. This site was a mean distance of 6 cm distal to the lateral epicondyle of the humerus. No compression of the PIN by the tendon of origin of the extensor carpi radialis brevis muscle was seen. One specimen was found to have a proximally split PIN that provided a previously undefined articular branch to the elbow joint. The mean diameter of the PIN proximal to the supinator muscle was 4.5 mm. The leash of Henry crossed the PIN in all but one specimen and was found at a mean distance of 5 cm inferior to the lateral epicondyle. The PIN exited the distal edge of the supinator muscle at a mean distance of 12 cm distal to the lateral epicondyle of the humerus. Here the mean diameter of the PIN was 4 mm. The exit site from the distal edge of the supinator was found to be at a mean distance of 18 cm proximal to the styloid process of the ulna. This exit site for the PIN was best identified following dissection between the extensor carpi radialis longus and brevis and extensor digitorum communis muscles. The distal articular branch of the PIN was found to have a mean length of 13 cm and the proximal portion of this terminal segment was located at a mean distance of 7.5 cm proximal to the Lister tubercle.
The addition of more anatomical landmarks can help the neurosurgeon to be more precise in identifying the PIN and in avoiding complications during surgery in this region.