20 Outcome of idiopathic normal pressure hydrocephalus
Adult Hydrocephalus, ed. Daniele Rigamonti. Published by Cambridge University Press. © Cambridge University Press 2014.
Introduction
Hakim et al. [1] introduced a unique triad of clinical symptoms defining normal pressure hydrocephalus (NPH), which included gait disturbance, dementia, and incontinence in association with enlarged ventricles and normal cerebrospinal fluid (CSF) pressure in his thesis “Some observations on CSF pressure: hydrocephalic syndrome in adults with ‘normal’ CSF pressure” and later in an article published by Hakim and Adams [2]. He reported that diversion of fluid using a shunt relieved the symptoms, and this has remained the primary treatment modality. Following shunting, many authors have reported that with appropriate diagnosis, the range of improvement varies from 24% to 90% [3–6].
However, studies had extreme ranges of follow-up periods from a few months to many years post-shunting in which long-term outcomes of shunted patients are few, partly due to the inherent loss of follow-up. The comorbidity which is a major confounder in outcome assessment in the elderly idiopathic normal pressure hydrocephalus (iNPH) patient population is seldom reported. There is a lack of standardized outcome scales, which makes such reporting difficult. The outcome results are graded based on clinical judgment by the physician as well as patient and caregiver opinions and are relatively subjective in nature and not quantified. Most importantly, there are no Class I data which address the best diagnostic strategy, surgical versus conservative management of iNPH, identifying one valve or configuration as superior to another, and reporting of shunt-related complications and malfunctions.
In a prospective study of 42 iNPH patients, Malm et al. [4] reported an overall improvement of 64% at 3 months compared with 26% at 3 years. At 3 months, 74% of his patients were independent, at 3 years the independent group was only 28%. Aygok et al. [5] reported from a prospective series of 50 iNPH patients that at 3-year follow-up, even though there was a slight decline, patients still retained a 75% improvement in gait. Moreover, 80% retained their improvement in memory and incontinence at 3 years post-shunt. These diverse findings are not only challenging but also are confusing for physicians, patients, and families when making diagnostic and treatment decisions. The following list represents the key steps which will hopefully serve as a guide to achieving a favorable outcome in iNPH patients:
proper selection criteria for diagnosis of iNPH;
deficiencies in identifying the degree of clinical impairment and therefore clinical improvement;
standardized management protocols related to outcome;
impact of complications after surgery;
impact of comorbidities.
Proper selection criteria for diagnosis of iNPH
Since the 1960s, great efforts have been made to improve diagnosis and management of NPH and achieve better selection of patients with idiopathic NPH (iNPH) and NPH of known cause (secondary NPH). Despite these efforts, there was a lack of Class I evidence for proper diagnosis and management, so the experts in the field assembled evidence-based guideline groups and published the first editions of international and Japanese guidelines for diagnosis and management of iNPH in 2005 and 2004, respectively, followed by the second edition of the Japanese guidelines in 2011 [7–13].
Based on these guideline recommendations, it is now essential to differentiate idiopathic NPH (iNPH) from other pathologies and to provide a multidimensional/multidisciplinary diagnostic approach which includes taking a detailed clinical history, assessing the triad of symptoms (gait/balance, cognition, urinary incontinence), and obtaining brain imaging [7,13,14]. Additionally, complementary diagnostic tests such as CSF infusion testing, CSF tap test or external lumbar drainage (ELD) when utilized in conjunction with clinical and radiological criteria, were found to identify iNPH patients more accurately and therefore better outcomes were achieved more efficiently [9,15].
In a prospective study of 43 iNPH patients, Walchenbach et al. [16] reported a positive predictive value of 87% with ELD for shunt outcome. Patients admitted to the Medical College of Virginia (MCV) who were diagnosed with iNPH according to a fixed protocol using diagnostic criteria based on gait disturbance, dementia, and incontinence in combination with ventriculomegaly without antecedent cause underwent a 3-day ELD procedure, and selection for surgery was made based on the improvement after ELD. The results indicated that ELD was the best prognostic indicator of a positive shunt outcome, with an accuracy of prediction greater than 90% [17]. The Study of Idiopathic Normal Pressure Hydrocephalus on Neurological Improvement: SINPHONI was conducted as a multicenter prospective study to evaluate the utility of MRI-based diagnosis for determining the 1-year outcome [18]. One hundred patients aged between 60 and 85 years with one or more iNPH symptoms (gait, cognitive, and urinary problems) and MRI evidence of disproportionately enlarged subarachnoid space hydrocephalus (DESH), i.e. tight high convexity and medial subarachnoid spaces and enlarged sylvian fissure associated with ventriculomegaly, received ventriculoperitoneal (VP) shunts with the Codman Hakim Programmable Valve. A favorable outcome was achieved in 69% of patients on the modified Rankin Scale whereas this rate was increased to 77% using the iNPH grading scale. The authors concluded that the MRI features of disproportionately enlarged subarachnoid space hydrocephalus (DESH) were useful for the diagnosis of iNPH. Advanced imaging modalities, measurement of regional cerebral blood flow (rCBF), acetazolamide challenge rCBF, and magnetic resonance spectroscopy (MRS) were also found to assist the selection of shunt responders [19].
It is clear that the increased chance of improvement after shunting was due to the proper selection of patients using dedicated clinical and radiological criteria and supplemental tests such as external lumbar drainage, which still remains the gold standard for predicting iNPH patients. Considering the fact that without surgery, most iNPH patients had measurable deterioration as early as 3 months following initial assessment, it is important to utilize standardized and validated criteria for proper diagnosis and conduct multicenter, randomized, placebo or cross-over designed clinical studies to obtain Class I evidence.
Deficiencies in identifying the degree of clinical impairment and therefore clinical improvement
Currently, there are neither standardized impairment nor standardized outcome scales that can be used for iNPH, such as the Glasgow Coma and Glasgow Outcome Scales that are used for head injury management [20]. Despite this fact, there are an increasing number of centers of excellence for iNPH management where clinicians have the opportunity to follow patients from baseline diagnosis to the plateau of improvement after surgery. These unique settings providing care for large numbers of iNPH patients enable clinicians to use and develop impairment and outcome scales for the management of iNPH even though validation and standardization of these tools is still limited (Table 20.1).
In an attempt to resolve the deficiencies in identifying the degree of clinical impairment in iNPH patients, the Kubo iNPH scale, Kiefer clinical grading, and iNPH scale (Tables 20.2–20.6) were introduced which measure impairment in the three major domains of iNPH symptoms: gait/balance, cognitive, and urinary continence [21– 23]. Furthermore, scales were developed to measure the impairment of individual iNPH symptoms only. For example, gait impairment can be measured using the iNPH MCV gait grade (Table 20.7), timed up and go test, and Tinetti tool, which all capture the key characteristics of iNPH gait pattern [17,24,25]. Cognitive impairment can be assessed with a commonly used outpatient test, the mini-mental state examination (MMSE), in addition to various neuropsychological tests of executive function (Stroop test), memory (Rey Auditory Verbal Learning Test [RAVLT]), and psychomotor function (GroovedPegboard [Lafayette Instrument Co., Lafayette, IN]). A depression scale such as the Beck Depression Inventory (BDI) can also be incorporated in the cognitive assessment [26–30]. For incontinence, validated questionnaires about urinary incontinence, overactive bladder, and quality of life related to urinary tract symptoms can be administered and, when necessary, urodynamics tests should be offered to obtain objective measures of urinary problems.
Table 20.2 Conversion of results on gait tests of the iNPH scale
The domain score is the mean value of the available converted scores.
Table 20.3 Conversion of results on neuropsychological tests of the iNPH scale
The domain score is the mean value of the available converted scores.
Table 20.4 Conversion of results on balance rating of the iNPH scale
Table 20.5 Conversion of results on continence rating of the iNPH scale
Table 20.6 Total score of the iNPH scale
Table 20.7 iNPH MCV gait grade
When used both pre- and postoperatively, these measures serve as a reliable tool in identifying the degree of clinical impairment and therefore the clinical improvement post-shunting. The difference between the preoperative and post-shunting follow-up scores on these scales has been able to provide an accurate means of outcome assessment. In addition, the increased certainty in diagnosing iNPH by using these dedicated scales has resolved the misinterpretation of atypical findings on imaging and challenged the common belief that very old age (above 80 years) has a negative impact on outcome. In a recent randomized study on hydrocephalus patients who had abundant white matter changes, which was generally thought to be a negative predictor for prognosis, the results indicated that these patients significantly improved after surgery if they were properly diagnosed [31]. Similarly, patients aged over 80 achieve a high degree of clinical improvement, despite the increased rate of comorbidities and complications that are mostly due to aging.
Standardized management protocols related to outcome
Since the publication of the international and Japanese guidelines, the establishment of common diagnostic criteria and measurement of characteristic features of iNPH have improved the management of iNPH patients [7,12]. Standardized protocols in the areas of shunt management, clinical outcome grading, time to follow-up after surgery, and clinical judgment are now under development which will enable comparability and reproducibility.
Shunt management
The pathophysiology of NPH secondary to other causes has been investigated in conjunction with the primary etiology. However, the pathophysiology of iNPH is still unclear, including the hypotheses of dysfunctional CSF absorption through arachnoid granulations or anomalies in the venous sinuses or iNPH being a “two hit” disease starting in infancy as “benign external hydrocephalus” followed by deep white matter ischemia (DWMI) in late adulthood [32]. Therefore, the purpose of surgical diversion of CSF via shunt in iNPH is to alleviate symptoms that have an unknown mechanism with high expectations of improvement after shunt surgery. In this regard, the exploration of the in vivo hydrodynamics of CSF shunts before making decisions between different shunt configurations and valve types and antisiphon devices plays a vital role.
Fixed differential pressure valves emerged as the first generation of treatment, to be followed by the introduction of flow-regulated and adjustable (programmable) valves. Adjustable valves have advantages for the management of improvement in triad symptoms and for resolving the problems of underdrainage and overdrainage noninvasively and thus better outcomes can be achieved without surgical intervention [10,33,34]. In the European iNPH study in which adjustable valves were implanted in all patients, one-third of the patients had shunt adjustments due to either over- or underdrainage [35]. In addition, standardized follow-up with control of shunt function and opening pressure in case of no or only partial improvement is important in order to achieve the benefit from shunting, as already pointed out by the same group and by various authors. Klinge et al. [35] reported from the same European iNPH study that 11% of their unimproved patients had “silent” shunt dysfunctions and were converted to improvers after noninvasive readjustments. A similar approach is also recommended in the Japanese guidelines [13] that for possible shunt dysfunction, two settings of reductions each for 30 mmH2O should be performed before considering a shunt revision. Furthermore, setting the postoperative shunt pressure to a higher setting and then gradually lowering the pressure after surgery will help elderly brains to cope better with the continuous CSF drainage through the shunt. An antisiphon device also serves as a siphon guard where excessive CSF flow may occur due to a positional siphoning effect.
In regard to shunt function, the standard approach is to obtain a cerebral image while exploring the in vivo CSF shunt dynamics. The imaging should be performed the day after surgery or just before discharge to serve as a baseline, and should be followed by further imaging 6 months to 1 year after surgery to show the effect of shunting. The type of imaging depends on the problem that needs to be identified. If an image is required for routine follow-up or if a shunt malfunction is suspected, a non-contrast head CT can be performed. However, if comorbidities such as dementia, atrophy, or organic etiologies need to be ruled out, then MRI is a better imaging modality. In addition, the utilization of routine shunt series should be considered when there is a suspected lack of physical integrity of the shunt or when there is a possibility of the distal end migration or abscess formation preceding overt shunt malfunction. Nevertheless, a CSF infusion test or a CSF tap test through the shunt chamber, even though they are relatively invasive in nature, provide the most accurate information in identifying the shunt malfunction [10–13,36].
In conclusion, it is now well understood that the shunt devices have a tremendous impact on the outcome of iNPH patients, therefore the valve type, configuration, opening and maintenance pressures, and auxiliary devices have to be properly selected with standard protocols. Any type of shunt malfunction has to be explored immediately by determining the best diagnostic approach available in that particular clinical setting. Only then, will the outcome in iNPH patients be accurately assessed and the rate of improvement after surgery be objectively determined.
Standardized outcome assessment
The commonly used outcome scales for iNPH, the modified Rankin Scale (mRS) and Barthel ADL Index (BI), measure the overall improvement or change in activities of daily living of people who mostly have suffered a stroke, which therefore makes it difficult to draw a clear picture for iNPH outcome [37–39]. In 1974, Stein and Langfitt [40] introduced an assessment of outcome in NPH patients based on activities of daily living – the Stein and Langfitt Scale. However, in the early studies, there was no consensus on which of these scales was most suitable for outcome assessment, therefore many authors decided to develop their own outcome measurement tool.
Boon et al. [41] used the Rankin Scale as a disability score and then developed a “Dutch NPH Scale.” Using Stein and Langfitt’s scale, Black et al. [42] reported an improvement in 33% of 62 iNPH patients. Simultaneously, they also assessed the improvement, which was graded based on the preoperative and postoperative changes in gait, dementia, and incontinence. They then created their own outcome scale using the above criteria and with their classification, the improvement rate was 50% in contrast to the 33% noted in the same group of patients using Stein and Langfitt’s scale. In the Medical College of Virginia, since 1992, preoperative and postoperative gait videos have been captured and analyzed to identify the impairment as well as the improvement of gait features in iNPH using the iNPH MCV gait grade presented in Table 20.7 [5,17]. For cognitive assessment, the mini-mental state examination (MMSE) and various neuropsychological tests were administered before and after surgery. For incontinence, the assessment was previously based on the patient’s and family’s own judgment. Since 2007, validated questionnaires about urinary incontinence, overactive bladder, and quality of life related to urinary tract symptoms have been administered before and after shunt treatment. If necessary, urodynamic tests were offered to obtain objective measures of urinary problems as well. A final conclusion about outcome was made based on the pre- and postoperative comparison of the triad symptoms using the above-mentioned assessments followed by classification of the overall improvement as “Excellent,” “Good,” “Partial,” or “None” using the Marmarou outcome scale [5].
More recently, a group of investigators developed a continuous, calibrated, and norm-based scale for grading the severity of symptoms and assessment of treatment outcome in iNPH [23]. This iNPH scale (Tables 20.2–20.6) was designed for the assessment of the four domains of gait, neuropsychology, balance, and continence, using ordinal ratings and continuous measures. The investigators reported that guidelines and papers discussing the prominent clinical features of iNPH, in conjunction with their clinical experience with iNPH patients, had served as the backbone for their development. In addition, data from normative studies were used to determine the limits for normal scores. Simultaneously, a prospective European multicentre study was conducted using this new iNPH scale as a diagnostic tool along with the radiological findings, and the primary outcome was measured as the difference between the preoperative and 12-month scores on the iNPH scale [35]. The presence of the four major domains (gait, neuropsychological, balance, and continence problems) in 142 patients with iNPH in the European multicenter study confirmed that this new iNPH scale provides proper diagnosis which distinguishes well between patients with different levels of symptom severity, and also between iNPH patients and healthy individuals. In regard to outcome, 84% of patients were improved one year after shunt treatment according to the new iNPH scale. In contrast, the change after surgery as measured by MRI showed an improvement in only 69% of these patients.
These findings are in agreement with the guideline recommendations that if ordinal and insensitive scales developed for other pathophysiological conditions are utilized to assess the outcome in iNPH patients, it will be impossible to accurately capture the improvement in the entire iNPH patient population. Therefore, preoperative and postoperative grading of the patient symptoms via gait analysis, neuropsychological testing, and internationally validated questionnaires for incontinence in conjunction with urodynamic studies are recommended for accurate diagnosis and outcome assessment. Other domains such as balance, dizziness, and headache scales can be incorporated if needed.
The time period for short-term and long-term outcome
The time for follow-up in iNPH patients had extreme ranges from months to years post-shunting, with short term referring to 3–6 months after surgery and long term to one year and more and even as long as 10–15 years post-surgery. For many years, investigators reported mixed results without identifying the specific time spans for follow-up. This resulted in various rates of improvement and therefore made it difficult to justify the long-term maintenance of favorable outcome after shunt surgery. Aygok et al. [5] reported that 3 months after shunt placement in 50 iNPH patients, 41% of patients had excellent/good recovery in gait whereas 50% had partial gait improvement. In general, overall gait improvement equaled 91%. For memory impairment, 40% of patients had excellent/good recovery and another 40% had partial recovery. For incontinence, 44% had improved dramatically and another 26% had partial improvement. At the end of 3 years, 92% of patients had survived. At three years, the improvement in gait was preserved for 75% of patients. Better results were seen in memory improvement at 3 years compared with 3 months; 80% of patients with memory impairment had sustained their recovery after 3 years and 20% had even reached a better improvement category. More than 80% of improvement was observed in incontinence in excellent/good/partial improvement categories.
The conclusion is that, if there is standardization of diagnosis and treatment and if all patients are being followed post-surgery within the same time intervals, then it is possible to see the positive effect of shunting in iNPH patients. Therefore, since the publication of the international and Japanese guidelines, there has been a consensus about obtaining regular follow-up assessments and reporting the time periods in which the outcome is assessed post-surgery [7,12]. Considering the fact that some iNPH patients have temporary improvements which are not long-lasting, 3–6 months post-surgery has to be considered as the short term and any time point after 6 months post-surgery has to be considered as the long-term follow-up.
Clinical judgment
The outcome results are graded based on clinical judgment performed by the physician since the investigators agree that this task mostly falls to the neurosurgeon, neurologist, geriatrician or sometimes the primary care physician. The patient and caregiver opinion is also being valued; however, it is important to remember that this could be relatively subjective in nature [43,44].
The standards regarding the frequency and nature of follow-up visits have now been established in the centers of excellence for iNPH management. Based on these suggestions, the neurosurgeon who performs the shunt operation should be seen within 2 weeks of surgery for suture removal and also for detecting any early complications. Then based on the type of implanted valve and the degree of improvement in iNPH symptoms, patients are scheduled for follow-up visits two to four times for the first year followed by bi-annual or annual visits afterwards. These assessments can be performed by a neurosurgeon, a neurologist, or a geriatrician and the consulting team should share the information with all the corresponding physicians who are involved in the patient’s care.
Impact of complications after surgery
Complications after shunting which impact the quality of clinical outcome include:
(a) major (subdural hematoma [SDH]/hygroma, shunt infection, and shunt failure); and
(b) minor (orthostatic headache, double vision, and hearing loss).
There is a strong correlation between outcome of shunted iNPH patients and complications after surgery. Even though shunting will eventually help to improve the symptoms of iNPH patients, complications if not treated will not only delay but also complicate the clinical improvement. Additionally, the different treatment options for various complications have major impacts on outcome post-shunting. The type of complication reversed noninvasively by resetting the shunt pressures mostly causes minor delays in clinical improvement. However, the surgical treatment of an SDH with a mass effect or a shunt removal for a serious shunt infection, recovery from which may take weeks to months, can cause major delays in achieving a favorable outcome. In a series of 50 iNPH patients treated with adjustable shunts, major (SDH, hygroma) and minor complications (headache, hearing loss, double vision) due to surgery totaled 6% and 14%, respectively [5]. All these patients improved without surgical intervention by adjusting the opening pressures. Hashimoto et al. (SINPHONI) [18] found there was a 3% incidence of serious events and 20% incidence of less serious events.
In conclusion, preventing complications during and after surgery before it occurs is crucial. To decrease the risk of over/underdrainage problems, valve and auxiliary device settings and configurations have to be properly selected based on the patient’s height, weight, and physical condition. For the risk of infection, the precautions have to be taken in the operating room and if infection is suspected then early detection is necessary. If adjustable valves are the choice of treatment, the goal is to have standardized protocols for managing the shunt opening pressures. One way is to set the pressure setting at a higher level during the shunt operation then to start reducing the opening pressure gradually. The same approach can be applied when treating the overdrainage complications: SDH, hygroma, headache, hearing loss, and double vision. Once these complications are resolved, then the shunt opening pressures can be gradually decreased. Nevertheless, if there is a serious risk such as mass effect and severe deterioration in clinical symptoms because of a complication, then patients have to be taken to the operating room for surgical intervention.
Impact of comorbidities
Comorbidity is a major confounder in outcome assessment in the iNPH patient population [45–47]. There is no doubt that health conditions like coexisting Parkinson’s disease, or spine, hip or knee problems in iNPH can affect the patient’s gait, and prostate problems may worsen the incontinence. Memory impairment can also be a part of vascular disease or Alzheimer’s disease.
Meier et al. [48] proposed a critical question in diagnosis and treatment of iNPH patients about whether preoperative factors can most reliably predict outcomes following shunt insertion. The results of the prospective follow-up examination (Kiefer Score, NPH Recovery Rate) of 82 out of 100 shunted iNPH patients were compared with the preoperative comorbidity index (CMI). Of the patients with a CMI score of 0–1, 67% experienced an excellent outcome, whereas a score of 4–5 was related to only 14% excellent. Remarkably, the excellent outcome of the patients between 6 and 8 on the CMI was 0%. The authors concluded that comorbidity is a statistically significant predictor of the clinical outcome for patients with iNPH undergoing shunt therapy. In a series reported by Aygok et al. [5], 24% (12/50) of the patients had comorbidities not related to surgery, including stroke, cardiac problems, vertigo, and psychiatric problems. Of these, stroke affected the independence in activities of daily living, whereas dementia and depression affected cognitive function, therefore improvement due to shunting became compromised. Since these comorbid conditions in the iNPH patient population have an influence on the clinical outcome after shunting, it is important to diagnose and treat these conditions.
Conclusion
With proper identification of iNPH patients and precise determination of the most suitable valve type and configuration, the improvement in outcome seen immediately after shunting is reasonably sustained over a period of years and provides a favorable risk/benefit profile of shunting in this elderly population. In addition, using standardized and validated iNPH scales that measure the clinical impairment before surgery and the clinical improvement after surgery will improve the outcome assessment during both short-term and long-term follow-up.
References
Findit@CUHK Library | Google Scholar
Findit@CUHK Library | Google Scholar | PubMed
Findit@CUHK Library | Google Scholar | PubMed
Findit@CUHK Library | Google Scholar | PubMedRelated posts:
25 – Chiari malformation and hydrocephalus in adults
11 – Core imaging in adult hydrocephalus
26 – Pseudotumor cerebri syndrome
22 – Low-pressure syndromes and cerebrospinal fluid leaks
24 – Management of hydrocephalus with associated cerebrospinal fluid pathologies
19 – Endoscopic third ventriculostomy
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
