Article Text

Download PDFPDF
Quantitative assessment of hand disability by the Nine-Hole-Peg test (9-HPT) in cervical spondylotic myelopathy
  1. S Olindo1,
  2. A Signate1,
  3. A Richech1,
  4. P Cabre1,
  5. Y Catonne2,
  6. D Smadja1,
  7. H Pascal-Mousselard2
  1. 1
    Department of Neurology, University Hospital Pierre Zobda-Quitman, Fort de France, Martinique, France
  2. 2
    Department of Orthopaedic Surgery, University Hospital Pitié-Salpétrière, University of Paris VI, Paris, France
  1. Stephane Olindo, MD, Department of Neurology, University Hospital Pierre Zobda-Quitman, 97261 Fort de France, Martinique, France; stephane.olindo{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Gait disturbances and sensory deficit in the hands are the most common clinical features in cervical spondylotic myelopathy (CSM). The disease course is often insidious and slow. The treatment for the underlying cervical-canal stenosis is surgical decompression or stabilisation. Functional scales have been used in CSM to assess neurological impairment, including the Nurick scale1 and the modified Japanese Orthopedic Association (mJOA) scale.2 Whereas functional scales are helpful in quantifying patient disability, they lack the objectivity of a neurological examination. A timed walking test as a quantitative functional test focusing on gait impairment has recently been developed.3 The present study has assessed hand disability in CSM patients with the quantitative Nine-Hole-Peg Test (9-HPT). This easy-to-use test was developed in 1985,4 and is now extensively applied to measure hand dexterity in multiple sclerosis (MS).5 Nine-HPT could constitute a quantifying complementary test to the Walking Time test in CSM to explore the two main symptoms—gait and hand disability.


Forty patients with CSM, corroborated by MRI, were recruited between January 2003 and March 2006. Patients with other known neurological disease were excluded from this study. A posterior decompressive procedure was undergone in all patients.

Twenty healthy subjects were assessed with the 9-HPT to give the normal range values in our population, defined as the mean time in seconds (s) ±2 standard deviations (SDs).

A measurement team composed of two neurologists underwent the clinical assessment in each hospital. A few days before the surgery, patients were examined and assessed by two neurologists. The assessments were repeated the day before and 1 month after the surgery.

All patients underwent the following functional tests: 9-HPT (s), Nurick test (0–5), mJOA (0–17), mJOA upper extremity sub-score (0–6) and 30 meters Walking Time test (s).

To simplify the task, we modified the 9-HPT. As previously described,4 patients were asked to take nine pegs from a bowl and then insert them into nine holes of a board as fast as possible. Patients were not asked to remove the pegs and to return them into the bowl. The test ended when all nine pegs were placed or after a maximum of 120 s. Patients had to complete the test two times. For each trial, the time was clocked in seconds and the faster trial was considered. The median 9-HPT score was also quantified and was defined as the average of dominant and non-dominant hands.

All computations were performed with the use of Statview (5.0) software.


The mean (SD) age and female/male sex ratio in 40 CSM patients and 20 control subjects were 63 (11.1) years and 17/23 (0.74), and 63.3 (13) years and 9/11 (0.81), respectively. Sex ratio and age were not statistically different between control subjects and CSM patients. Normal ranges for dominant and non-dominant hands were 9.2–22 s and 10.3–23.1 s. In pre-operative time, dominant and non-dominant 9-HPT were above the normal range limit in 17 (42.5%) and 20 (50%) patients, respectively, whereas in post-operative time, both hands were above the normal range limit in 10 (25%) patients. Mean and median 9-HPT in CSM patients in pre- and post-operative time and in control subjects were significantly different: 35±29.9 s and 22.2 s, 23.8±16.6 s and 18 s, 16.1±3.1 s and 16.2 s (p = 0.0005). Comparison of CSM test results in pre- and post-operative times are summarised in Table 1.

Table 1 Cervical spondylotic myelopathy test scores in pre- and post-operative time

All CSM test scores showed a statistically significant improvement in patients between pre- and post-operative times. In pre-operative time, the 9-HPT was not achieved in 120 s in two (5%) dominant hands and three (7.5%) non-dominant hands. In post-operative time, all patients achieved the 9-HPT in less than 120 s with both hands.

Median 9-HPT scores in pre-operative and post-operative time were significantly correlated with the other four tests. The poorer correlation was obtained with the Walking Time test (Table 2).

Table 2 Correlation between mean Nine-Hole-Peg Test and four measures of cervical spondylotic myelopathy in pre- and post-operative times

Intra-rater reliabilities were high, with coefficient correlations being estimated at 0.97 and 0.99 for dominant and non-dominant hands, respectively. Inter-rater agreements were also excellent, with coefficient correlations of 0.97 and 0.98 for dominant and non-dominant hands, respectively.


Hand disability in CSM is a complex symptom that associates motor weakness and sensory deficit in the upper limbs. Although several tests are used to assess neurological disability in CSM, none has been specifically developed to quantify finger dexterity. The Nurick scale1 and Walking Time test3 are based only on gait disability. The mJOA2 has subscales for upper extremity function with questions about hand-writing and use of utensils for eating. These functional scales are useful because they focus on the impairment of daily living activities. However, they lack the objectivity of a clinical neurological examination.

The aim of the present study was to propose, to clinicians and neurosurgeons, a quantitative test to measure the intensity of hand disability in CSM patients. The 9-HPT4 is a short and easily performed test and is one of the three components of the MS functional composite, which is a new outcome measure for MS clinical trials.5

The 9-HPT gives the opportunity to test dexterity hand by hand and reveals, in the present study, a high prevalence of hand disability in CSM patients (50% in pre-operative time).

Significant improvement in the month following surgery demonstrates the overall 9-HPT sensitivity to change as high as validated tests, such as the Walking Time test and mJOA scale. The 9-HPT is also characterised by high intra- and inter-rater reliability. It shows significant correlation with previously used scales, indicating validity and relevance. Although the feasibility of the 9-HPT appears excellent, patients with severe hand disability are sometimes unable to achieve the test. These conditions constitute a limitation of the use of the 9-HPT, as described in the methods of the present work.

In clinical practice, the 9-HPT seems to be an appropriate test to assess treatment efficiency in CSM. Although hand disability is not a recognised symptom of surgery indication, 9-HPT could constitute a precious tool in association with gait impairment assessment for treatment decisions in CSM. Two main clinical dimensions of CSM should be considered in any clinical trial outcome measure: ambulation and hand function. The 9-HPT may be highly suited to assess hand dexterity in CSM. These preliminary data represent a step in the development of a composite scale that could include the assessment of gait and hand motion disability by the Walking Time test and the 9-HPT.



  • Competing interests: None.