Date Published: June 19, 2018
Publisher: Public Library of Science
Author(s): Maria Kamenova, Jonathan Rychen, Raphael Guzman, Luigi Mariani, Jehuda Soleman, Benjamin Elder.
Despite being widely used, ventriculoperitoneal (VP) shunt placement is a procedure often associated with complications and revision surgeries. Many neurosurgical centers routinely perform early postoperative cranial computer tomography (CT) to detect postoperative complications (e.g., catheter malposition, postoperative bleed, over-drainage). Because guidelines are lacking, our study aimed to evaluate the yield of early routine postoperative CT after shunt placement for adult hydrocephalus. We retrospectively reviewed 173 patients who underwent frontal VP shunting for various neurosurgical conditions. Radiological outcomes were proximal catheter malposition, and ventricular width in preoperative and postoperative imaging. Clinical outcomes included postoperative neurological outcome, revision surgery because of catheter malposition or other causes, mortality, and finally surgical, non-surgical, and overall morbidity. In only 3 (1.7%) patients did the early routine postoperative CT lead to revision surgery. Diagnostic ratios for CT finding 1 asymptomatic patient who eventually underwent revision surgery per total number to scan were 1:58 for shunt malposition, 1:86 for hygroma, and 1:173 for a cranial bleed. Five (2.9%) patients with clinically asymptomatic shunt malposition or hygroma underwent intervention based on early postoperative CT (diagnostic ratio 1:25). Shunt malposition occurred in no patient with normal pressure hydrocephalus and 2 (40%) patients with stroke. Lower preoperative Evans’ Index was a statistically significant predictor for high-grade shunt malposition. We found a rather low yield for early routine postoperative cranial CT after frontal VP-shunt placement. Therefore, careful selection of patients who might benefit, considering the underlying disease and preoperative radiological findings, could reduce unnecessary costs and exposure to radiation.
Ventriculoperitoneal (VP) shunt placement is a standard procedure for the treatment of hydrocephalus for various conditions like normal pressure hydrocephalus (NPH), intracranial hemorrhage, aneurysmal subarachnoid hemorrhage (aSAH), meningitis, tumor, or trauma [1,2,3,4,5,6]. Despite being widely used, VP shunting is still plagued with complications that develop in 20 to 40% of the patients [7,8,9,10]. These often require revision surgeries that can increase morbidity and detrimentally affect quality of life [10,11,12].
We retrospectively reviewed 173 consecutive patients who underwent frontal VP shunt placement between February 2013 and June 2016 at our institution. The study protocol was approved by the local ethics committee (EKNZ, Basel, Switzerland). Diagnoses of symptomatic hydrocephalus of various underlying diseases were confirmed by cranial CT or magnetic resonance imaging (MRI). The VP shunt was placed through a frontal approach into the lateral horn aiming toward the foramen of Monro in all patients, except one who received a temporal shunt for the treatment of a trapped temporal horn. In all patients a programmable valve was implanted.
In our 173 consecutive patients with hydrocephalus caused by various diseases, we quantified a 1.7% yield for early postoperative imaging after VP shunting, specifically 3 patients underwent revision surgery for shunt malposition based solely on early routine postoperative CT. The ratio for finding 1 asymptomatic patient with proximal shunt malposition who eventually underwent revision surgery was 1:58 overall; 1:86 and 1:173 for hygroma due to over drainage (1.2%) or bleeding (0.6%), respectively; and 1:35 together for shunt malposition, hygroma, or bleeding. Lower preoperative EI was a factor statistically predictive for shunt malposition. A cutoff point of 0.285 showed a sensitivity of 85% and specificity of 67% for shunt malposition. Patients with shunt malposition had a significantly worse clinical outcome, and significantly higher rates of surgical morbidity, overall morbidity, and reoperation rates.
Our findings of low yields for routine early postoperative CT after frontal VP shunt placement address the paucity of information in the literature regarding the ideal timing for and yield of early imaging after shunting. Careful selection of the patients who might benefit based on the underlying disease and preoperative radiological findings might reduce unnecessary costs and exposure to radiation. With no NPH patient having an early CT scan that showed shunt malposition or overdrainage, imaging in this group might be redundant if clinical symptoms are lacking. In contrast, early postoperative CT might be reasonable or even recommended in patients with stroke, tumor, or meningitis when malposition is more common because of distorted cerebral anatomy.