Khoi D. Than, M.D., J. Nicole Bentley, M.D., and Anthony C. Wang, M.D.
Department of Neurosurgery, University of Michigan, Ann Arbor, Michigan, USA
Journal Club Article: Pinto, F, Saad, F, Oliveira, M, et al. Role of Endoscopic Third Ventriculostomy and Ventriculoperitoneal Shunt in Idiopathic Normal Pressure Hydrocephalus: Preliminary Results of a Randomized Clinical Trial. Neurosurgery. 2013; 72(5):845-854.
I. SIGNIFICANCE/CONTEXT AND IMPORTANCE OF THE STUDY
Normal pressure hydrocephalus (NPH) is a condition in which patients classically present with the clinical triad of cognitive decline, gait imbalance, and urinary incontinence. It is important to recognize this condition, as NPH represents one of just a few treatable dementias. Ventriculoperitoneal shunting (VPS) has met with good results, when offered judiciously. Although many patients have improvement in their symptoms after receiving this treatment, they often inherit the potential complications of VPS, specifically infection and malfunction. In this study, Pinto et al. report the results of a randomized, prospective clinical trial comparing the use of VPS and endoscopic third ventriculostomy (ETV) in patients with NPH.
II. ORIGINALITY OF THE WORK
This manuscript represents the first study to prospectively compare VPS and ETV in a randomized population of NPH patients.
III. APPROPRIATENESS OF STUDY DESIGN OR EXPERIMENTAL APPROACH
The most significant strength of this study was its randomized, prospective design. Additionally, 6 different outcomes scales were used, including the Mini-Mental Status Examination (MMSE), Berg Balance Scale (BBS), Functional Independence Measure (FIM), Dynamic Gait Index (DGI), NPH Japanese Scale, and Timed Up and Go (TUG) test.
IV. ADEQUACY OF EXPERIMENTAL TECHNIQUES
Although the experimental design was appropriate, there were 3 primary deficiencies in conduction of the study. As discussed in the paper, the tap test was used in patient selection, rather than lumbar puncture. The tap test has shown a high rate of false negative predictions in several series, and is not recommended as an exclusionary test due to low sensitivity (26-61%).1 Though specificity has been reported as fairly high, a better patient selection process may have improved upon the small sample size.
As a result of small sample size, there was a substantial discrepancy between the number of patients who underwent ETV (n = 12) compared to VPS (n = 26). With more than twice as many patients undergoing VPS than ETV, of whom 12 were in the “experimental” group, it is difficult to draw any firm conclusions from this comparison.
V. SOUNDNESS OF CONCLUSIONS AND INTERPRETATION
The primary outcome measure of this study was the NPH scale. Patients were considered improved if they had at least a 2-point reduction in NPH scale 1 year after treatment. The authors found that 77% of VPS patients improved compared to only 50% of ETV patients, a statistically significant difference. Analysis of Table 1, however, reveals that actual clinical improvement was probably subtle. ETV patients, on average, started with a worse clinical grade (NPH scale of 7) and improved by 1 point after 1 year. VPS patients started with an NPH scale of 6 and improved by an average of 2 points after 1 year. So, although 27% more patients with VPS had an “improvement” compared to ETV, the average benefit of just 1 point on a 12-point scale brings into question both the statistical and clinical significance of these findings.
Another factor that may bias the overall results is that there is no mention of any difference in medical comorbidities between the two groups, even though this data was collected. Clearly, if ETV patients were sicker as a whole, a smaller percentage of them would experience clinical improvement. Further, outcome measures calculated for the ETV group included 4 patients who required a subsequent procedure; all 4 patients who failed ETV were easy to identify as having failed the “flag sign.”
VI. RELEVANCE OF DISCUSSION
In the discussion, Pinto et al. address their study’s limitations, generalizability, and overall interpretation particularly well. However, there are two points of particular confusion. On page 850 they write, “…this study highlights the controversy regarding INPH treatment…” Currently, there is no controversy in the management of NPH, a fact which this study further supports; VPS is the established treatment, and ETV is being explored as a possible alternative. In addition, they also note, “The discrepancies between the results for patients less than 65 years of age compared with those over 65 years of age are intriguing.” Unfortunately, these discrepancies are not discussed elsewhere in the paper.
VII. CLARITY OF WRITING, STRENGTH AND ORGANIZATION OF THE PAPER
The authors have presented an important topic with a well-organized manuscript, although the writing could have been improved. On page 848, the authors write, “After the 3- and 12-month follow-up periods, 20 patients (20/26, 77%) showed improvement...In the remaining patients, ventricular reduction on the skull CT at 6 months was slight but evident. In these patients, the EI [Evans Index] ranged from 30% to 48.1% (mean, 38.1%) preoperatively, and 26.6% to 41.4% (mean, 36.6%) 6 months after VPS.” Here, it is unclear who the “remaining patients” refers to—presumably, the 6 patients who did not improve after VPS. Additionally, it is unclear why the same measurements were not assessed for the other patients. The reader must decipher whether the change in EI was more, less, or unchanged in that particular group. These values were also not included for the ETV group.
On page 849, Pinto et al. write, “…the TT [tap test] has a limited sensitivity (26%-61%) in comparison with the infusion test (57%-100%) and results in a prolonged external lumbar drainage in excess of 300 mL (50%-100%), which may cause missing of potential patients.” The wording of this sentence is in error, but its meaning is evident.
Also on page 849, the authors write, “Because gait apraxia was the only clinical outcome with a significant difference between treatment groups, there may be a potential diagnosis bias because of the limited values of the MMSE.” This sentence is flawed for a number of reasons. First, the reader is left to assume that “gait apraxia” refers to the DGI scale and not the TUG test or the gait subsection on the NPH scale. Second, nowhere in the article were the statistical differences between the two treatment groups’ performance on each outcome measure actually reported or discussed. Third, it is very unclear how the second part of the sentence relates to the first.
Further critique of the manuscript reveals several important omissions in addition to the ones already discussed. For example, it is not discussed whether there was a difference in duration of symptoms between the two groups. It is also not discussed whether any of the 30 patients who underwent VPS developed shunt failure requiring reoperation at any point during that year of follow-up.
VIII. ECONOMY OF WORDS
The authors’ use of words was perhaps too economical. In two instances (pp. 847 and 852) the authors write that they “…strongly believe that a patient with a previous ETV who is then submitted to a VPS has a very different clinical profile compared with a patient with no previous ETV.” While this is probably true, it is also dependent on how long the ETV was present, other anatomical considerations, etc. Pinto et al. could have expounded upon this belief with further explanation or support from the literature.
IX. RELEVANCE, ACCURACY AND COMPLETENESS OF BIBLIOGRAPHY
Two articles pertaining to ETV for NPH were absent from the bibliography. In 2011, Cage et al.2 did compare outcomes in NPH patients who underwent VPS versus ETV. These authors found that a higher percentage of ETV patients (89%) did not require in-home assistance compared to VPS patients (54%). A higher percentage of VPS patients (72%), however, reported improvement in quality of life compared to ETV patients (56%). In 2012, Fountas et al.3 prospectively studied 7 patients who underwent ETV for NPH. This study found an average improvement in NPH scale from 6.1 to 3.1 and MMSE from 20.6 to 23.6.
In addition, on page 849 in defending the use of the tap test over other supplemental diagnostic techniques, the authors cite only 2 papers, one of which compares the tap test to lumbar drainage, and the other evaluates lumbar drainage separately. While they use a range of values in each point, these ranges are not within the results of the papers that they cite.
X. FUTURE/NEXT STEPS
Future studies will need to enroll more patients, have more evenly sized groups, and be conducted at multiple centers in order to truly demonstrate that VPS is better than ETV for the treatment of NPH.
The authors would like to thank Holly Wagner for her editorial assistance in the preparation of this manuscript.
1. Marmarou A, Bergsneider M, Klinge P, Relkin N, Black PM. The value of supplemental prognostic tests for the preoperative assessment of idiopathic normal-pressure hydrocephalus. Neurosurgery. 2005;57(3 Suppl):ii-v.
2. Cage TA, Auguste KI, Wrensch M, Wu YW, Gupta N. Self-reported functional outcome after surgical intervention in patients with idiopathic normal pressure hydrocephalus. J Clin Neurosci. 2011;18(5):649-654.
3. Fountas KN, Kapsalaki EZ, Paterakis KN, Lee GP, Hadjigeorgiou GM. Role of endoscopic third ventriculostomy in treatment of selected patients with normal pressure hydrocephalus. Acta Neurochir Suppl. 2012;113:129-133.