Pituitary DOI 10.1007/s11102-015-0642-9
Median nerve conduction studies and wrist magnetic resonance imaging in acromegalic patients with carpal tunnel syndrome Yasuo Sasagawa • Osamu Tachibana • Mariko Doai • Hisao Tonami • Hideaki Iizuka
Ó Springer Science+Business Media New York 2015
Abstract Purpose Carpal tunnel syndrome (CTS) often occurs with acromegaly; however, the pathophysiology of CTS in acromegalic patients remains unclear. This study evaluated the median nerve in acromegalic patients with and without CTS. Methods We examined the median nerves of 21 acromegalic patients (eight patients with CTS and 13 patients without CTS) using electrophysiological nerve conduction studies and wrist magnetic resonance images. They underwent transsphenoidal surgery to resect their growth hormone-secreting pituitary adenomas. The median nerves of the patients with CTS were reassessed by the same studies. Results The sensory conduction velocity was significantly later in the median nerves of patients with CTS than in patients without CTS (34.9 vs. 45.8 m/s, respectively; P = 0.006). In the wrist magnetic resonance images, the cross-sectional area of the median nerve in CTS patients and non-CTS patients was 18.7 and 10.5 mm2, respectively. The median nerve was significantly larger in patients with CTS than in patients without CTS (P \ 0.003). The flattering ratio of the median nerve and palmar deviation of the flexor retinaculum were not significantly different between the two patient groups. After tumor resection, the nerve conduction velocities improved
Y. Sasagawa (&) O. Tachibana H. Iizuka Department of Neurosurgery, Kanazawa Medical University, 1-1 Daigaku, Uchinada 920-0293, Ishikawa, Japan e-mail: [email protected] M. Doai H. Tonami Department of Diagnostic and Therapeutic Radiology, Kanazawa Medical University, Uchinada, Ishikawa, Japan
in patients with CTS, but the nerve remained enlarged. The CTS symptoms disappeared in all patients, except one. Conclusions The median nerves of acromegalic patients with CTS were enlarged and had impaired nerve conduction. This finding represents a predominant intrinsic feature in the pathophysiology of the disease rather than an extrinsic feature such as a thickened transverse carpal ligament. Keywords Carpal tunnel syndrome Median nerve Acromegaly Nerve conduction study Wrist magnetic resonance image
Introduction Median nerve entrapment in the wrist to palm segment produces the clinical condition carpal tunnel syndrome (CTS). It is the most common entrapment neuropathy. Carpal tunnel syndrome reportedly occurs in 34–64 % of acromegalic patients [1–3], and is often the initial complaint. However, from a pathophysiological point of view, there is no consensus in literature to explain the nerve involvement in acromegaly. Previous studies of median neuropathy in acromegaly have proposed various pathogenic mechanisms such as increased connective tissue within the carpal tunnel [4], bony or synovial overgrowth of the carpal bones [5], or increased extracellular fluid within the tunnel [6]. A diagnosis of CTS has typically relied on clinical features and electrophysiological data. In a recent radiological study, the median nerve was easily visible on axial magnetic resonance (MR) image as a round or flattened structure of intermediate signal intensity. Reported MR image findings with idiopathic CTS in patients without
123
Pituitary
acromegaly include severe flattening of the median nerve within the carpal tunnel, increased signal intensity on T2weighted images because of increased edema, and palmar bowing of the flexor retinaculum [7–9]. Only a few studies have analyzed the median nerve of acromegalic patients by using electrophysiological data or MR images [1, 10, 11]. This study evaluated the median nerve in acromegalic patients with CTS in comparison to asymptomatic acromegalic patients. The pathophysiology of the median nerve of CTS in acromegaly is also discussed.
Methods Patient population Forty-five newly diagnosed acromegalic patients were treated at Kanazawa Medical University Hospital (Ishikawa, Japan) between May 2005 and August 2014. The diagnosis of acromegaly was based on established criteria [12]. Twenty-one patients consented to participate in this study. For statistical analysis, the patients were divided into the CTS group and the non-CTS group (i.e., patients without CTS). The patients in the CTS group complained of sensory disturbances (e.g., palmar numbness and paresthesia) in the territory of distribution of the median nerve. All studied patients underwent transsphenoidal surgery (which was performed by the second author [O. T.]) to resect growth hormone (GH)-secreting pituitary adenomas. Standard histological examination of paraffin-embedded tissue sections and immunohistochemical studies confirmed the diagnosis of GH-secreting pituitary adenoma in all patients. Postoperative disease activity was evaluated 3 months to 1 year after the surgery. The evaluation was based on the serum GH measurement during the oral glucose tolerance test and on the basal value of insulin-like growth factor 1 (IGF-1), which was calculated as the standard deviation (SD). Electrophysiological study Before surgery, electrophysiological studies were performed on symptomatic hands in the CTS group and on all hands in the non-CTS group. Clinical neurophysiology specialists measured the nerve conduction velocity by electromyography (Neuropack MEB-2200 electromyograph; Nihon-Kohden, Tokyo, Japan). The temperature of the extremity was monitored and maintained constantly above 32 °C. The motor nerve conduction velocity (MCV) was measured using a recording electrode placed at the center of the abductor pollicis brevis muscle. Stimuli were applied to the wrist joint 70 mm proximal to the recording electrodes. Distal motor latency (DML) was also measured.
123
Retrograde sensory nerve conduction velocity (SCV) was measured by placing a recording ring electrode at the base of the middle finger. Based on a past study [13], the normal limits of the SCV, MCV, and DML were defined, as follows: SCV, 45.5 m/s; MCV, 42.5 m/s; and DML, 4.4 m/s. Wrist MR image Magnetic resonance images were obtained using a 1.5 T scanner (Magnetom Avanto; Siemens, Munich, Germany) with a four-channel flex small coil. All wrists were imaged in the neutral position with the fingers extended. The MR image investigators were blinded to clinical and electrophysiological findings. Fast spin-echo T1-weighted sequence and T2-weighted sequence were used. The imaging parameters of the axial T1-weighted images were slice thickness, 3 mm; field of view (FOV), 100 mm; matrix, 256 9 230; number of excitations (NEX), 1; echo train length (ETL), 2; repetition time (TR), 465 ms; and echo time (TE), 11 ms. The imaging parameters of the axial and coronal T2-weighted images were slice thickness, 2 mm; FOV, 150 mm; matrix, 256 9 256; NEX, 2; ETL, 12; TR, 4,000 ms; and TE, 95 ms. All available images were interpreted independently by two observers. The following variables were assessed at the level of the hook of the hamate bone on T2 axial image: (1) the cross-sectional area of median nerve; the area was calculated by the formula: (major axis/ 2) 9 (minor axis/2) 9 p (circular constant); (2) the flattering ratio of the median nerve; the ratio was calculated by the formula: major axis/minor axis; (3) the high-intensity signal of the median nerve; the values were corrected, based on the intensity of the surrounding fat to avoid variations in signal that are encountered with surface coils; and (4) the palmar deviation of the flexor retinaculum; to quantify the palmar deviation of the flexor retinaculum, a straight line was drawn that united the trapezium bone and the hook of the hamate bone. The palmar deviation was measured as the distance of this line to the vertex of bowing of flexor retinaculum. Figure 1 shows the transverse illustration and the T2weighted axial image of the wrist at the level of the hook of the hamate bone and the contents of carpal tunnel. Statistical analysis The data are presented as the mean (±SD) for the parametric data and as the median (±SD) for the nonparametric data. Statistical analyses were performed using Excel 2010 (Microsoft, Redmond, WA, USA) add-in software Ekuseru-Toukei 2010 for Windows version (Social Survey
Pituitary Fig. 1 a Transverse illustration and b T2-weighted axial image of the wrist at the level of the hook of the hamate bone and the contents of the carpal tunnel
Research Information, Tokyo, Japan). The Mann–Whitney U test and Wilcoxon signed-rank test were performed for the unpaired and paired continuous data, respectively. Fisher’s extract test was performed for categorical data.
Results Patient characteristics Eight (38.1 %) of 21 patients complained of sensory disturbances (e.g., palmar numbness and paresthesia) in the territory of distribution of the median nerve. These eight patients were the CTS group. In this group, four patients had CTS symptoms in the bilateral hands and the remaining patients had CTS symptoms in the unilateral hand. Four patients were men and four patients were women. The mean age was 54.9 years. The levels of serum GH and IGF-1, and the SD of IGF-1 value of each age group in the CTS patients were 22.3 ng/mL, 638.6 ng/mL, and 6.97 SD, respectively. The mean glycated hemoglobin (hemoglobin A1c) level of the CTS group was 6.6 %. There were no statistical differences in baseline characteristics between the CTS group and the non-CTS group (Table 1).
Electrophysiological evaluations before surgery Electrophysiological studies were evaluated on the symptomatic hands (n = 12) of the CTS group patients and on all hands (n = 26) in the non-CTS group patients. Table 2 shows the comparisons of electrophysiological data between acromegalic patients with and without CTS. The mean SCV, MCV, and DML in the CTS group were 34.9 m/s, 49.4 m/s, and 5.7 ms, respectively. The mean SCV and DML in the CTS group were below the normal limits. Nine (75 %) of 12 hands in the CTS group had a later SCV than the normal limits. The nerves of the CTS group had a significantly later SCV, compared to the nerves of the non-CTS group (34.9 vs. 45.8 m/s, respectively; P = 0.006). There were no statistically significant differences between the two groups in the MCV and DML. In the non-CTS group, approximately one-half of the median nerves [11 (42.3 %) of 26 hands] had a later SCV than the normal limits. Wrist MR findings before surgery Twelve wrists of symptomatic hands in the CTS group and 26 wrists in the non-CTS group were evaluated. Table 3
Table 1 Baseline characteristics of acromegalic patients with and without carpal tunnel syndrome Characteristics
Acromegalic patients
P
With CTS (n = 8)
Without CTS (n = 13)
Male/female ratio
4/4
5/8
0.221
Symptomatic hand
Unilateral hand, 4 pts bilateral hands, 4 pts
–
–
Age (years)
54.9 ± 10.1
61.2 ± 8.9
0.261
GH (ng/mL)
22.3 ± 11.4
12.7 ± 6.2
0.156
IGF-1 (ng/mL)
638.6 ± 190.2
586.7 ± 166.8
0.625
SD of IGF-1
6.97 ± 2.4
6.59 ± 2.6
0.114
HbA1c (%)
6.6 ± 0.9
6.4 ± 0.7
0.883
The data are presented as the mean ± SD CTS carpal tunnnel syndrome, GH growth hormone, HbA1c glycated hemoglobin, IGF-1 insulin-like growth factor-1, pts patients, SD standard deviation
123
Pituitary Table 2 Comparison of the median nerve electrophysiological data between acromegalic patients with and without carpal tunnel syndrome
SCV (m/s)
With CTS (n = 12)
Without CTS (n = 26)
P
34.9 ± 10.4
45.8 ± 8.8
0.006*
49.4 ± 20.2
55.8 ± 3.7
0.074
5.7 ± 2.4
3.9 ± 0.8
0.093
(normal limit, \45.5) MCV (m/s) (normal limit, \42.5) DML (ms) (normal limit, \4.4) The data are presented as the mean ± SD CTS carpal tunnel syndrome, DML distal motor latency, MCV motor conduction velocity, SCV sensory conduction velocity, SD standard deviation * P \ 0.01
Table 3 Comparison of quantitative wrist magnetic resonance image data between acromegaly patients with and without carpal tunnel syndrome With CTS (n = 12)
Without CTS (n = 26)
P
Cross-sectional area (mm2)
18.7 ± 3.7
10.5 ± 4.3
0.003*
Flattening ratio
2.2 ± 0.5
2.3 ± 0.4
0.31
High-intensity signal
6/12
10/26
0.72
Palmar deviation (mm)
2.2 ± 0.4
1.6 ± 0.6
0.14
Unless otherwise indicated, the data are presented as the mean ± SD CTS carpal tunnel syndrome, SD standard deviation * P \ 0.01
Fig. 2 T2-weighted axial images at the level of the hook of the hamate show a the enlarged median nerve (crosssectional area, 20.3 mm2) of an acromegalic patient with CTS and b the median nerve (crosssectional area, 7.9 mm2) of a patient without CTS
shows the comparisons of the quantitative wrist MR image data between acromegalic patients with and without CTS. The mean cross-sectional area of the median nerve in the CTS group and non-CTS group were 18.7 and 10.5 mm2, respectively. The CTS group patients had a significantly larger median nerve, compared to the non-CTS group
123
patients (P \ 0.003) (Fig. 2). The T2 weighted images showed high-intensity signals in six of 10 hands in the CTS group and in 10 of 26 hands in the non-CTS group (P = 0.72). In the flattering ratio and palmar deviation, there were no significant correlations between the CTS and non-CTS groups.
Pituitary Table 4 Comparison of preoperative and postoperative data of median nerve electrophysiological studies and wrist magnetic resonance images of acromegalic patients with carpal tunnel syndrome (n = 12) Preoperative
Postoperative
P
SCV (m/s)
34.9 ± 10.4
48.1 ± 4.6
0.008**
MCV (m/s)
49.4 ± 20.2
52.7 ± 3.5
0.07
DML (ms) Cross-sectional area (mm2)
5.7 ± 2.4 18.7 ± 3.7
3.5 ± 0.9 14.2 ± 1.9
0.03* 0.08
Flattening ratio
2.2 ± 0.5
2.1 ± 0.8
0.31
High-intensity signal
6/12
4/12
0.34
Palmar deviation (mm)
2.2 ± 0.4
1.8 ± 0.8
0.29
Unless otherwise indicated, the data are presented as the mean ± SD CTS carpal tunnel syndrome, DML distal motor latency, MCV motor conduction velocity, SCV sensory conduction velocity, SD standard deviation * P \ 0.05 ** P \ 0.01
Postoperative course in the CTS group Among the eight patients in the CTS group, transsphenoidal surgery cured six patients of GH hypersecretion. Two patients who were not cured by surgery received adjuvant medical therapy and continued their treatment with cabergoline. The IGF-1 value in all patients who received medical therapy were controlled at less than ?2 SD. Three months to 1 year after the surgery, the eight patients in the CTS group underwent the same electrophysiological studies and wrist MR image. Table 4 shows a comparison of the pre- and postoperative data of the median nerve electrophysiological studies and wrist MR images in the acromegalic patients with CTS. In the electrophysiological studies, the median nerve had a significantly faster SCV (P = 0.008) and DML (P = 0.03) after surgery, compared to their preoperative values. The MCV was also faster after surgery than before surgery, although there was no statistical difference. In the wrist MR images, the high-intensity signal in the median nerves remained after surgery. In the cross-sectional area, the flattering ratio of the median nerves and palmar deviation were also unchanged after surgery. The CTS symptoms immediately improved after surgery in all patients, except one. This patient (a 60-year-old woman) had ongoing finger numbness on her left hand, although her serum GH and IGF-1 returned to a normal range after the tumor resection. This symptom was intractable, and she underwent a carpal tunnel release operation 1 year after the transsphenoidal surgery. Discussion The results of this study showed that the median nerve of acromegalic patients with CTS has delayed conduction
velocity and is enlarged. We found that the nerve conduction impairment improved after the GH level reduced after treatment. Eight (38.1 %) of 21 patients with acromegaly complained of sensory disturbances in the territory of distribution of the median nerve in our study. The finding of CTS in acromegaly was consistent with 34–64 % reported in studies [1–3]. This study showed no significant difference between the CTS and non-CTS groups in the baseline values (i.e., GH and IGF-1). The small sample size may have also resulted in the lack of difference in the baseline values. The disease duration of acromegaly may be an important factor in developing CTS, although it is actually difficult to know the precise duration of acromegaly. The SCV and DML of the median nerve in the CTS group were below the normal limits. The SCVs were significantly slower in the CTS group than in the non-CTS group. Electrophysiological tests are established in the diagnosis of CTS. The SCV is one of the most useful parameters with 85–95 % sensitivity for the early diagnosis of CTS [14, 15]. Baum et al. [1] conducted an electrophysiological study with 50 acromegalic patients. Twentyeight (56 %) of these patients had an abnormal limit in distal sensory latency (i.e., greater than 4.5 ms). Kameyama et al. [10] also studied median nerve conduction in 16 acromegalic patients with asymptomatic CTS. Thirteen (81 %) of the 16 asymptomatic patients had electrophysiological evidence of subclinical CTS. In our study, approximately 40 % of the non-CTS group patients had a SCV below the normal limit and a high-intensity signal in the median nerve on MR images. Nerve conduction impairments in acromegalic patients exist for quite a while before CTS symptoms manifest. Jenkins et al. [11] evaluated the median nerves of nine acromegalic patients using wrist MR images. Four patients with CTS symptoms had an enlarged nerve (based on the cross-sectional area) and signal intensity, compared to the five asymptomatic patients; however, the two groups did not differ in the volume of carpal tunnel contents (e.g., the degree of palmar bowing of the flexor retinaculum). Our data is consistent with this previous study. Furthermore, Tagliafico et al. [3] examined the median nerve and ulnar nerve of 34 acromegalic patients with ultrasound. Their study included a relatively large cohort and 34 normal controls. The nerves of acromegalic patients had a significantly greater cross-sectional area in the median nerve and ulnar nerve, compared to healthy controls. Among the acromegalic patients, patients with uncontrolled disease had a greater cross-sectional area in the median and ulnar nerves, compared to disease-controlled patients. They concluded that peripheral nerve enlargement in acromegaly may be an intrinsic feature of the disease with regard to clinical control, disease duration, and IGF-1 levels. Based
123
Pituitary
on the past and our studies, the predominant pathology of CTS in acromegaly would be increased edema of the median nerve within the carpal tunnel. Median neuropathy seems to be induced by nerve conduction impairment after nerve enlargement at the carpal tunnel level rather than by compressive neuropathy (e.g., increased connective tissue, bony or synovial overgrowth). Nerve enlargement in acromegalic patients would be a direct effect of IGF-1. This is in line with data in the literature showing that IGF-1 directly influences nerve tissue [16, 17]. Insulin-like growth factor 1 affects multiple layers of the regenerative response in peripheral nerve regeneration, and it promotes nerve elongation and branching [17]. In the postoperative course of the CTS group, CTS symptoms disappeared in all patients, except one. Electrophysiological studies demonstrated an improvement of conduction impairments. However, the cross-sectional area or T2 high-intensity signals of the median nerve did not change. It may be that these changes in MR imaging take a long time to manifest. By contrast, one patient had ongoing finger numbness on her hand and underwent carpal tunnel release surgery. Iwasaki et al. [18] described the surgical efficacy of carpal tunnel release for CTS in four cases of acromegaly. Before performing carpal tunnel release, these four patients received tumor resection or medical control for GH-secreting pituitary adenoma. However, the patients had ongoing CTS symptoms, despite controlling GH secretion. At the operation, all patients had a thickened transverse carpal ligament. The median nerve was flattened beneath the incised ligament in all patients. These findings (e.g., thickened transverse carpal ligament or flattened nerve) generally occur in idiopathic CTS. In some acromegalic patients, intrinsic and extrinsic factors may intermingle in the pathology of median neuropathy. Nevertheless, definitive conclusions on the pathophysiology of CTS in acromegaly need more sufficient cases that compare the nerve of acromegalic patients with idiopathic CTS patients and healthy controls.
Conclusion Our study suggests that the median nerves of acromegalic patients with CTS are significantly enlarged and have impaired conduction, compared to the nerves of non-CTS patients. This finding seems to represent an intrinsic feature of the disease. The rapid improvement of impaired nerve conduction after pituitary hypersecretion treatment, which is associated with the resolution of neuropathic symptoms, indicates that the early diagnosis of acromegaly and prompt control of GH and IGF-1 levels may enable nerve normalization and reduce the clinical manifestation of acromegalic neuropathy.
123
Acknowledgments The authors received no financial support in conjunction with the generation of this submission. Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Baum H, Lu¨decke DK, Herrmann HD (1986) Carpal tunnel syndrome and acromegaly. Acta Neurochir (Wien) 83(1–2):54–55 2. O’Duffy JD, Randall RV, MacCarty CS (1973) Median neuropathy (carpal-tunnel syndrome) in acromegaly. A sign of endocrine overactivity. Ann Intern Med 78:379–383 3. Tagliafico A, Resmini E, Nizzo R, Bianchi F, Minuto F, Ferone D et al (2008) Ultrasound measurement of median and ulnar nerve cross-sectional area in acromegaly. J Clin Endocrinol Metab 93:905–909 Epub 2007 Dec 11 4. Schiller F, Kolb FO (1954) Carpal tunnel syndrome in acromegaly. Neurology 4:271–282 5. Gondring WH (1966) The carpal tunnel syndrome and acromegaly. J Okla State Med Assoc 59:274–279 6. Johnston AW (1960) Acroparaesthesiae and acromegaly. Br Med J 1:1616–1618 7. Tsujii M, Hirata H, Morita A, Uchida A (2009) Palmar bowing of the flexor retinaculum on wrist MRI correlates with subjective reports of pain in carpal tunnel syndrome. J Magn Reson Imaging 29:1102–1105 8. Cudlip SA, Howe FA, Clifton A, Schwartz MS, Bell BA (2002) Magnetic resonance neurography studies of the median nerve before and after carpal tunnel decompression. J Neurosurg 96:1046–1051 9. Horch RE, Allmann KH, Laubenberger J, Langer M, Stark GB (1997) Median nerve compression can be detected by magnetic resonance imaging of the carpal tunnel. Neurosurgery. 41:76–82 (discussion 82–83) 10. Kameyama S, Tanaka R, Hasegawa A, Tamura T, Kuroki M (1993) Subclinical carpal tunnel syndrome in acromegaly. Neurol Med Chir (Tokyo) 33:547–551 11. Jenkins PJ, Sohaib SA, Akker S, Phillips RR, Spillane K, Wass JA et al (2000) The pathology of median neuropathy in acromegaly. Ann Intern Med 133:197–201 12. Giustina A, Barkan A, Casanueva FF, Cavagnini F, Frohman L, Ho K et al (2000) Criteria for cure of acromegaly: a consensus statement. J Clin Endocrinol Metab 85:526–529 13. Chang MH, Liu LH, Lee YC, Wei SJ, Chiang HL, Hsieh PF (2006) Comparison of sensitivity of transcarpal median motor conduction velocity and conventional conduction techniques in electrodiagnosis of carpal tunnel syndrome. Clin Neurophysiol 117:984–991 14. Sharma KR, Rotta F, Romano J, Ayyar DR (2001) Early diagnosis of carpal tunnel syndrome: comparison of digit 1 with wrist and distoproximal ratio. Neurol Clin Neurophysiol 2001:2–10 15. Dudley Porras AF, Rojo Alaminos P, Vinuales JI, Ruiz Villamanan MA (2000) Value of electrodiagnostic tests in carpal tunnel syndrome. J Hand Surg Br 25:361–365 16. Rabinovsky ED (2004) The multifunctional role of IGF-1 in peripheral nerve regeneration. Neurol Res 26:204–210 17. Mohammadi R, Saadati A (2014) Influence of insulin-like growth factor I on nerve regeneration using allografts: a sciatic nerve model. J Craniofac Surg 25:1510–1514 18. Iwasaki N, Masuko T, Ishikawa J, Minami A (2005) Surgical efficacy of carpal tunnel release for carpal tunnel syndrome in acromegaly: report of four patients. J Hand Surg Br 30:605–606
Median nerve conduction studies and wrist magnetic resonance imaging in acromegalic patients with carpal tunnel syndrome Yasuo Sasagawa • Osamu Tachibana • Mariko Doai • Hisao Tonami • Hideaki Iizuka
Ó Springer Science+Business Media New York 2015
Abstract Purpose Carpal tunnel syndrome (CTS) often occurs with acromegaly; however, the pathophysiology of CTS in acromegalic patients remains unclear. This study evaluated the median nerve in acromegalic patients with and without CTS. Methods We examined the median nerves of 21 acromegalic patients (eight patients with CTS and 13 patients without CTS) using electrophysiological nerve conduction studies and wrist magnetic resonance images. They underwent transsphenoidal surgery to resect their growth hormone-secreting pituitary adenomas. The median nerves of the patients with CTS were reassessed by the same studies. Results The sensory conduction velocity was significantly later in the median nerves of patients with CTS than in patients without CTS (34.9 vs. 45.8 m/s, respectively; P = 0.006). In the wrist magnetic resonance images, the cross-sectional area of the median nerve in CTS patients and non-CTS patients was 18.7 and 10.5 mm2, respectively. The median nerve was significantly larger in patients with CTS than in patients without CTS (P \ 0.003). The flattering ratio of the median nerve and palmar deviation of the flexor retinaculum were not significantly different between the two patient groups. After tumor resection, the nerve conduction velocities improved
Y. Sasagawa (&) O. Tachibana H. Iizuka Department of Neurosurgery, Kanazawa Medical University, 1-1 Daigaku, Uchinada 920-0293, Ishikawa, Japan e-mail: [email protected] M. Doai H. Tonami Department of Diagnostic and Therapeutic Radiology, Kanazawa Medical University, Uchinada, Ishikawa, Japan
in patients with CTS, but the nerve remained enlarged. The CTS symptoms disappeared in all patients, except one. Conclusions The median nerves of acromegalic patients with CTS were enlarged and had impaired nerve conduction. This finding represents a predominant intrinsic feature in the pathophysiology of the disease rather than an extrinsic feature such as a thickened transverse carpal ligament. Keywords Carpal tunnel syndrome Median nerve Acromegaly Nerve conduction study Wrist magnetic resonance image
Introduction Median nerve entrapment in the wrist to palm segment produces the clinical condition carpal tunnel syndrome (CTS). It is the most common entrapment neuropathy. Carpal tunnel syndrome reportedly occurs in 34–64 % of acromegalic patients [1–3], and is often the initial complaint. However, from a pathophysiological point of view, there is no consensus in literature to explain the nerve involvement in acromegaly. Previous studies of median neuropathy in acromegaly have proposed various pathogenic mechanisms such as increased connective tissue within the carpal tunnel [4], bony or synovial overgrowth of the carpal bones [5], or increased extracellular fluid within the tunnel [6]. A diagnosis of CTS has typically relied on clinical features and electrophysiological data. In a recent radiological study, the median nerve was easily visible on axial magnetic resonance (MR) image as a round or flattened structure of intermediate signal intensity. Reported MR image findings with idiopathic CTS in patients without
123
Pituitary
acromegaly include severe flattening of the median nerve within the carpal tunnel, increased signal intensity on T2weighted images because of increased edema, and palmar bowing of the flexor retinaculum [7–9]. Only a few studies have analyzed the median nerve of acromegalic patients by using electrophysiological data or MR images [1, 10, 11]. This study evaluated the median nerve in acromegalic patients with CTS in comparison to asymptomatic acromegalic patients. The pathophysiology of the median nerve of CTS in acromegaly is also discussed.
Methods Patient population Forty-five newly diagnosed acromegalic patients were treated at Kanazawa Medical University Hospital (Ishikawa, Japan) between May 2005 and August 2014. The diagnosis of acromegaly was based on established criteria [12]. Twenty-one patients consented to participate in this study. For statistical analysis, the patients were divided into the CTS group and the non-CTS group (i.e., patients without CTS). The patients in the CTS group complained of sensory disturbances (e.g., palmar numbness and paresthesia) in the territory of distribution of the median nerve. All studied patients underwent transsphenoidal surgery (which was performed by the second author [O. T.]) to resect growth hormone (GH)-secreting pituitary adenomas. Standard histological examination of paraffin-embedded tissue sections and immunohistochemical studies confirmed the diagnosis of GH-secreting pituitary adenoma in all patients. Postoperative disease activity was evaluated 3 months to 1 year after the surgery. The evaluation was based on the serum GH measurement during the oral glucose tolerance test and on the basal value of insulin-like growth factor 1 (IGF-1), which was calculated as the standard deviation (SD). Electrophysiological study Before surgery, electrophysiological studies were performed on symptomatic hands in the CTS group and on all hands in the non-CTS group. Clinical neurophysiology specialists measured the nerve conduction velocity by electromyography (Neuropack MEB-2200 electromyograph; Nihon-Kohden, Tokyo, Japan). The temperature of the extremity was monitored and maintained constantly above 32 °C. The motor nerve conduction velocity (MCV) was measured using a recording electrode placed at the center of the abductor pollicis brevis muscle. Stimuli were applied to the wrist joint 70 mm proximal to the recording electrodes. Distal motor latency (DML) was also measured.
123
Retrograde sensory nerve conduction velocity (SCV) was measured by placing a recording ring electrode at the base of the middle finger. Based on a past study [13], the normal limits of the SCV, MCV, and DML were defined, as follows: SCV, 45.5 m/s; MCV, 42.5 m/s; and DML, 4.4 m/s. Wrist MR image Magnetic resonance images were obtained using a 1.5 T scanner (Magnetom Avanto; Siemens, Munich, Germany) with a four-channel flex small coil. All wrists were imaged in the neutral position with the fingers extended. The MR image investigators were blinded to clinical and electrophysiological findings. Fast spin-echo T1-weighted sequence and T2-weighted sequence were used. The imaging parameters of the axial T1-weighted images were slice thickness, 3 mm; field of view (FOV), 100 mm; matrix, 256 9 230; number of excitations (NEX), 1; echo train length (ETL), 2; repetition time (TR), 465 ms; and echo time (TE), 11 ms. The imaging parameters of the axial and coronal T2-weighted images were slice thickness, 2 mm; FOV, 150 mm; matrix, 256 9 256; NEX, 2; ETL, 12; TR, 4,000 ms; and TE, 95 ms. All available images were interpreted independently by two observers. The following variables were assessed at the level of the hook of the hamate bone on T2 axial image: (1) the cross-sectional area of median nerve; the area was calculated by the formula: (major axis/ 2) 9 (minor axis/2) 9 p (circular constant); (2) the flattering ratio of the median nerve; the ratio was calculated by the formula: major axis/minor axis; (3) the high-intensity signal of the median nerve; the values were corrected, based on the intensity of the surrounding fat to avoid variations in signal that are encountered with surface coils; and (4) the palmar deviation of the flexor retinaculum; to quantify the palmar deviation of the flexor retinaculum, a straight line was drawn that united the trapezium bone and the hook of the hamate bone. The palmar deviation was measured as the distance of this line to the vertex of bowing of flexor retinaculum. Figure 1 shows the transverse illustration and the T2weighted axial image of the wrist at the level of the hook of the hamate bone and the contents of carpal tunnel. Statistical analysis The data are presented as the mean (±SD) for the parametric data and as the median (±SD) for the nonparametric data. Statistical analyses were performed using Excel 2010 (Microsoft, Redmond, WA, USA) add-in software Ekuseru-Toukei 2010 for Windows version (Social Survey
Pituitary Fig. 1 a Transverse illustration and b T2-weighted axial image of the wrist at the level of the hook of the hamate bone and the contents of the carpal tunnel
Research Information, Tokyo, Japan). The Mann–Whitney U test and Wilcoxon signed-rank test were performed for the unpaired and paired continuous data, respectively. Fisher’s extract test was performed for categorical data.
Results Patient characteristics Eight (38.1 %) of 21 patients complained of sensory disturbances (e.g., palmar numbness and paresthesia) in the territory of distribution of the median nerve. These eight patients were the CTS group. In this group, four patients had CTS symptoms in the bilateral hands and the remaining patients had CTS symptoms in the unilateral hand. Four patients were men and four patients were women. The mean age was 54.9 years. The levels of serum GH and IGF-1, and the SD of IGF-1 value of each age group in the CTS patients were 22.3 ng/mL, 638.6 ng/mL, and 6.97 SD, respectively. The mean glycated hemoglobin (hemoglobin A1c) level of the CTS group was 6.6 %. There were no statistical differences in baseline characteristics between the CTS group and the non-CTS group (Table 1).
Electrophysiological evaluations before surgery Electrophysiological studies were evaluated on the symptomatic hands (n = 12) of the CTS group patients and on all hands (n = 26) in the non-CTS group patients. Table 2 shows the comparisons of electrophysiological data between acromegalic patients with and without CTS. The mean SCV, MCV, and DML in the CTS group were 34.9 m/s, 49.4 m/s, and 5.7 ms, respectively. The mean SCV and DML in the CTS group were below the normal limits. Nine (75 %) of 12 hands in the CTS group had a later SCV than the normal limits. The nerves of the CTS group had a significantly later SCV, compared to the nerves of the non-CTS group (34.9 vs. 45.8 m/s, respectively; P = 0.006). There were no statistically significant differences between the two groups in the MCV and DML. In the non-CTS group, approximately one-half of the median nerves [11 (42.3 %) of 26 hands] had a later SCV than the normal limits. Wrist MR findings before surgery Twelve wrists of symptomatic hands in the CTS group and 26 wrists in the non-CTS group were evaluated. Table 3
Table 1 Baseline characteristics of acromegalic patients with and without carpal tunnel syndrome Characteristics
Acromegalic patients
P
With CTS (n = 8)
Without CTS (n = 13)
Male/female ratio
4/4
5/8
0.221
Symptomatic hand
Unilateral hand, 4 pts bilateral hands, 4 pts
–
–
Age (years)
54.9 ± 10.1
61.2 ± 8.9
0.261
GH (ng/mL)
22.3 ± 11.4
12.7 ± 6.2
0.156
IGF-1 (ng/mL)
638.6 ± 190.2
586.7 ± 166.8
0.625
SD of IGF-1
6.97 ± 2.4
6.59 ± 2.6
0.114
HbA1c (%)
6.6 ± 0.9
6.4 ± 0.7
0.883
The data are presented as the mean ± SD CTS carpal tunnnel syndrome, GH growth hormone, HbA1c glycated hemoglobin, IGF-1 insulin-like growth factor-1, pts patients, SD standard deviation
123
Pituitary Table 2 Comparison of the median nerve electrophysiological data between acromegalic patients with and without carpal tunnel syndrome
SCV (m/s)
With CTS (n = 12)
Without CTS (n = 26)
P
34.9 ± 10.4
45.8 ± 8.8
0.006*
49.4 ± 20.2
55.8 ± 3.7
0.074
5.7 ± 2.4
3.9 ± 0.8
0.093
(normal limit, \45.5) MCV (m/s) (normal limit, \42.5) DML (ms) (normal limit, \4.4) The data are presented as the mean ± SD CTS carpal tunnel syndrome, DML distal motor latency, MCV motor conduction velocity, SCV sensory conduction velocity, SD standard deviation * P \ 0.01
Table 3 Comparison of quantitative wrist magnetic resonance image data between acromegaly patients with and without carpal tunnel syndrome With CTS (n = 12)
Without CTS (n = 26)
P
Cross-sectional area (mm2)
18.7 ± 3.7
10.5 ± 4.3
0.003*
Flattening ratio
2.2 ± 0.5
2.3 ± 0.4
0.31
High-intensity signal
6/12
10/26
0.72
Palmar deviation (mm)
2.2 ± 0.4
1.6 ± 0.6
0.14
Unless otherwise indicated, the data are presented as the mean ± SD CTS carpal tunnel syndrome, SD standard deviation * P \ 0.01
Fig. 2 T2-weighted axial images at the level of the hook of the hamate show a the enlarged median nerve (crosssectional area, 20.3 mm2) of an acromegalic patient with CTS and b the median nerve (crosssectional area, 7.9 mm2) of a patient without CTS
shows the comparisons of the quantitative wrist MR image data between acromegalic patients with and without CTS. The mean cross-sectional area of the median nerve in the CTS group and non-CTS group were 18.7 and 10.5 mm2, respectively. The CTS group patients had a significantly larger median nerve, compared to the non-CTS group
123
patients (P \ 0.003) (Fig. 2). The T2 weighted images showed high-intensity signals in six of 10 hands in the CTS group and in 10 of 26 hands in the non-CTS group (P = 0.72). In the flattering ratio and palmar deviation, there were no significant correlations between the CTS and non-CTS groups.
Pituitary Table 4 Comparison of preoperative and postoperative data of median nerve electrophysiological studies and wrist magnetic resonance images of acromegalic patients with carpal tunnel syndrome (n = 12) Preoperative
Postoperative
P
SCV (m/s)
34.9 ± 10.4
48.1 ± 4.6
0.008**
MCV (m/s)
49.4 ± 20.2
52.7 ± 3.5
0.07
DML (ms) Cross-sectional area (mm2)
5.7 ± 2.4 18.7 ± 3.7
3.5 ± 0.9 14.2 ± 1.9
0.03* 0.08
Flattening ratio
2.2 ± 0.5
2.1 ± 0.8
0.31
High-intensity signal
6/12
4/12
0.34
Palmar deviation (mm)
2.2 ± 0.4
1.8 ± 0.8
0.29
Unless otherwise indicated, the data are presented as the mean ± SD CTS carpal tunnel syndrome, DML distal motor latency, MCV motor conduction velocity, SCV sensory conduction velocity, SD standard deviation * P \ 0.05 ** P \ 0.01
Postoperative course in the CTS group Among the eight patients in the CTS group, transsphenoidal surgery cured six patients of GH hypersecretion. Two patients who were not cured by surgery received adjuvant medical therapy and continued their treatment with cabergoline. The IGF-1 value in all patients who received medical therapy were controlled at less than ?2 SD. Three months to 1 year after the surgery, the eight patients in the CTS group underwent the same electrophysiological studies and wrist MR image. Table 4 shows a comparison of the pre- and postoperative data of the median nerve electrophysiological studies and wrist MR images in the acromegalic patients with CTS. In the electrophysiological studies, the median nerve had a significantly faster SCV (P = 0.008) and DML (P = 0.03) after surgery, compared to their preoperative values. The MCV was also faster after surgery than before surgery, although there was no statistical difference. In the wrist MR images, the high-intensity signal in the median nerves remained after surgery. In the cross-sectional area, the flattering ratio of the median nerves and palmar deviation were also unchanged after surgery. The CTS symptoms immediately improved after surgery in all patients, except one. This patient (a 60-year-old woman) had ongoing finger numbness on her left hand, although her serum GH and IGF-1 returned to a normal range after the tumor resection. This symptom was intractable, and she underwent a carpal tunnel release operation 1 year after the transsphenoidal surgery. Discussion The results of this study showed that the median nerve of acromegalic patients with CTS has delayed conduction
velocity and is enlarged. We found that the nerve conduction impairment improved after the GH level reduced after treatment. Eight (38.1 %) of 21 patients with acromegaly complained of sensory disturbances in the territory of distribution of the median nerve in our study. The finding of CTS in acromegaly was consistent with 34–64 % reported in studies [1–3]. This study showed no significant difference between the CTS and non-CTS groups in the baseline values (i.e., GH and IGF-1). The small sample size may have also resulted in the lack of difference in the baseline values. The disease duration of acromegaly may be an important factor in developing CTS, although it is actually difficult to know the precise duration of acromegaly. The SCV and DML of the median nerve in the CTS group were below the normal limits. The SCVs were significantly slower in the CTS group than in the non-CTS group. Electrophysiological tests are established in the diagnosis of CTS. The SCV is one of the most useful parameters with 85–95 % sensitivity for the early diagnosis of CTS [14, 15]. Baum et al. [1] conducted an electrophysiological study with 50 acromegalic patients. Twentyeight (56 %) of these patients had an abnormal limit in distal sensory latency (i.e., greater than 4.5 ms). Kameyama et al. [10] also studied median nerve conduction in 16 acromegalic patients with asymptomatic CTS. Thirteen (81 %) of the 16 asymptomatic patients had electrophysiological evidence of subclinical CTS. In our study, approximately 40 % of the non-CTS group patients had a SCV below the normal limit and a high-intensity signal in the median nerve on MR images. Nerve conduction impairments in acromegalic patients exist for quite a while before CTS symptoms manifest. Jenkins et al. [11] evaluated the median nerves of nine acromegalic patients using wrist MR images. Four patients with CTS symptoms had an enlarged nerve (based on the cross-sectional area) and signal intensity, compared to the five asymptomatic patients; however, the two groups did not differ in the volume of carpal tunnel contents (e.g., the degree of palmar bowing of the flexor retinaculum). Our data is consistent with this previous study. Furthermore, Tagliafico et al. [3] examined the median nerve and ulnar nerve of 34 acromegalic patients with ultrasound. Their study included a relatively large cohort and 34 normal controls. The nerves of acromegalic patients had a significantly greater cross-sectional area in the median nerve and ulnar nerve, compared to healthy controls. Among the acromegalic patients, patients with uncontrolled disease had a greater cross-sectional area in the median and ulnar nerves, compared to disease-controlled patients. They concluded that peripheral nerve enlargement in acromegaly may be an intrinsic feature of the disease with regard to clinical control, disease duration, and IGF-1 levels. Based
123
Pituitary
on the past and our studies, the predominant pathology of CTS in acromegaly would be increased edema of the median nerve within the carpal tunnel. Median neuropathy seems to be induced by nerve conduction impairment after nerve enlargement at the carpal tunnel level rather than by compressive neuropathy (e.g., increased connective tissue, bony or synovial overgrowth). Nerve enlargement in acromegalic patients would be a direct effect of IGF-1. This is in line with data in the literature showing that IGF-1 directly influences nerve tissue [16, 17]. Insulin-like growth factor 1 affects multiple layers of the regenerative response in peripheral nerve regeneration, and it promotes nerve elongation and branching [17]. In the postoperative course of the CTS group, CTS symptoms disappeared in all patients, except one. Electrophysiological studies demonstrated an improvement of conduction impairments. However, the cross-sectional area or T2 high-intensity signals of the median nerve did not change. It may be that these changes in MR imaging take a long time to manifest. By contrast, one patient had ongoing finger numbness on her hand and underwent carpal tunnel release surgery. Iwasaki et al. [18] described the surgical efficacy of carpal tunnel release for CTS in four cases of acromegaly. Before performing carpal tunnel release, these four patients received tumor resection or medical control for GH-secreting pituitary adenoma. However, the patients had ongoing CTS symptoms, despite controlling GH secretion. At the operation, all patients had a thickened transverse carpal ligament. The median nerve was flattened beneath the incised ligament in all patients. These findings (e.g., thickened transverse carpal ligament or flattened nerve) generally occur in idiopathic CTS. In some acromegalic patients, intrinsic and extrinsic factors may intermingle in the pathology of median neuropathy. Nevertheless, definitive conclusions on the pathophysiology of CTS in acromegaly need more sufficient cases that compare the nerve of acromegalic patients with idiopathic CTS patients and healthy controls.
Conclusion Our study suggests that the median nerves of acromegalic patients with CTS are significantly enlarged and have impaired conduction, compared to the nerves of non-CTS patients. This finding seems to represent an intrinsic feature of the disease. The rapid improvement of impaired nerve conduction after pituitary hypersecretion treatment, which is associated with the resolution of neuropathic symptoms, indicates that the early diagnosis of acromegaly and prompt control of GH and IGF-1 levels may enable nerve normalization and reduce the clinical manifestation of acromegalic neuropathy.
123
Acknowledgments The authors received no financial support in conjunction with the generation of this submission. Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Baum H, Lu¨decke DK, Herrmann HD (1986) Carpal tunnel syndrome and acromegaly. Acta Neurochir (Wien) 83(1–2):54–55 2. O’Duffy JD, Randall RV, MacCarty CS (1973) Median neuropathy (carpal-tunnel syndrome) in acromegaly. A sign of endocrine overactivity. Ann Intern Med 78:379–383 3. Tagliafico A, Resmini E, Nizzo R, Bianchi F, Minuto F, Ferone D et al (2008) Ultrasound measurement of median and ulnar nerve cross-sectional area in acromegaly. J Clin Endocrinol Metab 93:905–909 Epub 2007 Dec 11 4. Schiller F, Kolb FO (1954) Carpal tunnel syndrome in acromegaly. Neurology 4:271–282 5. Gondring WH (1966) The carpal tunnel syndrome and acromegaly. J Okla State Med Assoc 59:274–279 6. Johnston AW (1960) Acroparaesthesiae and acromegaly. Br Med J 1:1616–1618 7. Tsujii M, Hirata H, Morita A, Uchida A (2009) Palmar bowing of the flexor retinaculum on wrist MRI correlates with subjective reports of pain in carpal tunnel syndrome. J Magn Reson Imaging 29:1102–1105 8. Cudlip SA, Howe FA, Clifton A, Schwartz MS, Bell BA (2002) Magnetic resonance neurography studies of the median nerve before and after carpal tunnel decompression. J Neurosurg 96:1046–1051 9. Horch RE, Allmann KH, Laubenberger J, Langer M, Stark GB (1997) Median nerve compression can be detected by magnetic resonance imaging of the carpal tunnel. Neurosurgery. 41:76–82 (discussion 82–83) 10. Kameyama S, Tanaka R, Hasegawa A, Tamura T, Kuroki M (1993) Subclinical carpal tunnel syndrome in acromegaly. Neurol Med Chir (Tokyo) 33:547–551 11. Jenkins PJ, Sohaib SA, Akker S, Phillips RR, Spillane K, Wass JA et al (2000) The pathology of median neuropathy in acromegaly. Ann Intern Med 133:197–201 12. Giustina A, Barkan A, Casanueva FF, Cavagnini F, Frohman L, Ho K et al (2000) Criteria for cure of acromegaly: a consensus statement. J Clin Endocrinol Metab 85:526–529 13. Chang MH, Liu LH, Lee YC, Wei SJ, Chiang HL, Hsieh PF (2006) Comparison of sensitivity of transcarpal median motor conduction velocity and conventional conduction techniques in electrodiagnosis of carpal tunnel syndrome. Clin Neurophysiol 117:984–991 14. Sharma KR, Rotta F, Romano J, Ayyar DR (2001) Early diagnosis of carpal tunnel syndrome: comparison of digit 1 with wrist and distoproximal ratio. Neurol Clin Neurophysiol 2001:2–10 15. Dudley Porras AF, Rojo Alaminos P, Vinuales JI, Ruiz Villamanan MA (2000) Value of electrodiagnostic tests in carpal tunnel syndrome. J Hand Surg Br 25:361–365 16. Rabinovsky ED (2004) The multifunctional role of IGF-1 in peripheral nerve regeneration. Neurol Res 26:204–210 17. Mohammadi R, Saadati A (2014) Influence of insulin-like growth factor I on nerve regeneration using allografts: a sciatic nerve model. J Craniofac Surg 25:1510–1514 18. Iwasaki N, Masuko T, Ishikawa J, Minami A (2005) Surgical efficacy of carpal tunnel release for carpal tunnel syndrome in acromegaly: report of four patients. J Hand Surg Br 30:605–606
