Clinical outcomes of nonsurgical treatment for Preiser disease.

To elucidate whether nonsurgical treatment for Preiser disease is effective.Eight patients with Preiser disease (median age 59 [47-69] years) underwent nonsurgical treatment (median symptom-onset-to-treatment interval 8 [9-180] months). At presentation, 7 patients complained of constant pain and 1 of motion-related pain. Pain restricted wrist range of motion (median modified Mayo wrist score [MMWS] 17.5 [range 10-30]). Radiography revealed stages 1 to 3 disease (Herbert-Lanzetta classification). Median scapholunate angle was 62° (54°-75°), with 3 wrists suffering dorsal intercalated segment instability (DISI). Magnetic resonance imaging showed (Kalainov criteria) 4 stage 1 wrists (complete necrosis) and 4 stage 2 (incomplete necrosis). Two had concomitant Kienböck disease. All patients underwent nonsurgical treatment (ie, oral pain killer, immobilization, rest) and were monitored via radiographic and clinical evaluations. Scapholunate angles and the scaphoid area reduction ratio were calculated using radiography. Response criteria were the patients' subjective and objective status. Endpoint was the time from start of non-surgical to surgical treatment.Immobilization lasting 0 to 24 months (median 1.8 months) did not relieve their symptoms. Follow-up radiography showed that the disease stage had progressed in 5 of 8 wrists, with 5 wrists having DISI. The median area reduction ratio of the scaphoid was 11% (4%-52%) on anteroposterior views and 4% (-23% to 17%) on lateral views. Compared with the contralateral wrist, the median wrist flexion-extension arc was 61% (50%-79%) and the median grip strength 39%. Median MMWS score was 17.5 (10-25) - poor in 6 of 8 patients. Surgery was thus necessary in all patients.Nonsurgical treatment for Preiser disease did not improve subjective or objective outcomes and did not prevent deterioration of radiographic findings.Type of study/level of evidence: Therapeutic, Level V.

Introduction

Preiser disease – idiopathic avascular necrosis (AVN) of the scaphoid – is an uncommon disorder that was first reported by Preiser in 1910.[ AVN of the scaphoid has been reported in patients with no significant history of trauma but who complained of increasing, severe radial wrist pain with spontaneous onset. Physical examination showed tenderness around the dorsoradial aspect of the wrist and swelling around the wrist joint. To date, although the etiology and pathology of this disease have not been established, idiopathic vascular insufficiency of the carpals and repetitive mechanical stress on the wrist are regarded as its cause.[ Most cases have involved systemic steroid use[ for a systemic illness, such as autoimmune hemolytic anemia,[ systemic lupus erythematosus,[ or renal transplantation.[ Another etiology could be topical steroid use for a local illness such as de Quervain disease.[. In addition, not a few cases have involved risk factors such as smoking, alcoholism, and/or infection.[

Although both nonsurgical and surgical treatment for Preiser disease have been reported, there is no consensus on the optimal treatment because of the rarity of the disease.[ Some researchers reported favorable clinical outcomes of nonsurgical treatment for Preiser disease in a few cases.[ Others; however, reported that nonsurgical treatment (eg, immobilization, oral analgesics, corticosteroid injections, electrical stimulation) have not provided satisfactory outcomes.[ Thus, whether nonsurgical treatment for Preiser disease is effective is a highly controversial issue. The purpose of this study was to investigate the clinical outcomes of 8 patients with Preiser disease treated non-surgically to elucidate whether such an effort is effective.

Materials and methods

This retrospective single-centered consecutive case series was performed at our institution. This study was conducted in conformity with the ethical guidelines of the 1975 Declaration of Helsinki after approval from our institutional review board, the ethic committee of Nippon Medical School Hospital (No. 30-07-970). For this retrospective investigation, patients’ demographic characteristics, medical history, imaging findings, and follow-up data were extracted from their clinical records and medical linkage systems.

Patients

From April 1998 to March 2015, 8 patients (1 man, 7 women) with Preiser disease were treated in our hospital. The demographic data for the patients at the first visit in our hospital are shown in Table 1. All patients, with no significant history of trauma, had complained of increasing, intolerable dorsoradial wrist pain of spontaneous onset. There was no history of any steroid use in 6 of the patients. One patient had been given a total of 8 steroid injections (2 mg; betamethasone sodium phosphate; Rinderon) in the contralateral subacromial bursa to treat periarthritis of the left shoulder during the 5 to 8 years before the definitive diagnosis of AVN of the scaphoid in the right wrist. Another patient had 5 injections of betamethasone phosphate (2 mg; Rinderon), and 4 injections of triamcinolone acetonide (10 mg) around a painful wrist joint during the 6 years before the current definitive diagnosis.[ The physical examination showed tenderness around the dorsoradial aspect of the wrist and swelling around the wrist joint.

Table 1

Patients’ demographic data at the initial visit to our hospital.

All patients were examined with plain radiography, and 3 patients underwent computed tomography as well. AVN of the scaphoid was diagnosed by radiographic evidence (ie, sclerosis, fragmentation, erosion, collapse of the scaphoid in the absence of a significant history of trauma). Magnetic resonance imaging (MRI) was performed in all patients, and evidence of AVN on MRI was confirmed by radiologists and an orthopedic surgeon. Osteonecrosis was defined as a region of bone marrow with low-intensity signal on both T1-weighted imaging (T1WI) and T2WI. Ischemic changes or bone edema were identified as areas of bone marrow with patchy, decreased, or absent signal on T1WI when compared with the surrounding carpal bones but with patchy, isointense, or high signal on T2WI. Contrast-enhanced MRI was performed to gain a definitive diagnosis in a patient, and showed the decrease in bone marrow perfusion.

Eight patients with Preiser disease who underwent non-surgical treatment were a median age of 59 years (range 47–69 years) at the time of their first visit. The period from the onset of symptoms to the start of nonsurgical treatment was a median 8 months (range 9–180 months). At the time of the first visit, 7 patients complained of constant pain, and 1 complained of motion-related pain. The radiographic and MRI results, the wrists’ range of motion (ROM), and grip strength at the first visit to our hospital are shown in Table 2. The data for pain duration and the modified Mayo wrist score (MMWS)[ are shown in Table 3. Staging classification of the disease and clinical outcomes (MMWS) are shown in Table 4 and Table 5, respectively. One wrist had stage 1 disease, 4 had stage 2, 3 had stage 3 (Herbert–Lanzetta classification). The median scapholunate angle (SLA) was 62° (range 54°–75°), which indicated that 3 wrists had dorsal intercalated segment instability (DISI). Plain radiography and/or computed tomography showed that 3 patients had stage 1 osteoarthritis of the radioscaphoid joint according to the Watson classification.[ MRI showed that 4 wrists had stage 1 (Kalainov criteria) (complete necrosis), and 4 had stage 2 (incomplete necrosis). Two wrists had concomitant Kienböck disease. The ROMs of the wrists were restricted by pain, and the median MMWS was 17.5 points (range 10–30 points).

Table 2

Data for the patients at the initial visit to our hospital.

Table 3

Further data of the patients at the initial visit to our hospital.

Table 4

Staging classification of avascular necrosis of the scaphoid and lunate and classification of osteoarthritis on the wrist joint according to radiography and MRI.

Table 5

Modified Mayo wrist score.

The response criteria were defined as the subjective and objective status at the clinical and radiological evaluations, respectively, and the endpoint was defined as the time from starting nonsurgical treatment to the surgical intervention to mitigate the intolerable wrist pain. All patients underwent a follow-up examination that included radiographic evaluation, interviews regarding wrist pain, and evaluation of ROM, grip strength, and MMWS.

Informed consent was obtained in writing from all patients for the nonsurgical treatment and publication of the cases at the first visit our hospital.

Nonsurgical treatment

Non-surgical treatment – including oral administration of nonsteroidal anti-inflammatory drugs, immobilization with a splint or orthosis, and rest – was undertaken for various periods.

Radiographic evaluation

The SLA was calculated on lateral radiographs. The DISI deformity was evaluated according to the SLA (>70°). To measure the whole area of the scaphoid, the area of the scaphoid was traced and calculated using image analysis software Leafareacounter Plus (https://www.vector.co.jp/download/file/win95/business/fh625135.html) on anteroposterior and lateral radiographs. The area reduction ratio of the scaphoid was presented as a percentile – that is, dividing the area of the scaphoid at the endpoint by that on the initial radiographs, subtract from 1, and multiply by 100.

Clinical evaluation

The clinical evaluation included ROM, the wrist flexion-extension arc, grip strength measured with a dynamometer (Jamar; Baran/Tec, Clifton, NJ), and calculation of the MMWS. We then calculated the ratio of the wrist flexion–extension arc of the affected wrist compared that of the unaffected wrist. Also, the ratio of the grasp strength of the affected wrist compared that of the unaffected wrist with a correction for limb dominance was also evaluated. This correction was always conducted when the dominant hand was the affected hand. The grip strength of the dominant hand was assumed to be 15% stronger than the nondominant hand. A correction was not conducted for the nondominant hand.

Results

The results of the nonsurgical treatment are shown in Table 6 and Table 7. The median duration of immobilization was 1.8 months (range 0–24 months). There were no complications (eg, neuropathy, contracture of fingers). Our patients were initially treated nonsurgically (ie, wrist immobilization with a splint or orthosis) for various periods. To mitigate the pain, all patients took painkillers, including loxoprofen (180 mg), lornoxicam (24 mg), or mofezolac (225 mg) per a day. The drugs were administered in the full therapeutic dose. The wrist pain; however, was not alleviated in any of the patients, indicating that the nonsurgical treatment was ineffective. Surgery was; therefore, ultimately required for all patients.

Table 6

Data of the patients at the endpoint.

Table 7

Further data of the patients at the endpoint.

Radiographic evaluations

The relationship between Herbert–Lanzetta classification and the follow-up periods are shown in Figure 1. Follow-up radiographs and the Herbert–Lanzetta classification[ at the endpoint showed that 2 wrists had stage 2 disease, 4 had stage 3, and 2 had stage 4. Thus, 2 of the wrists had progressed to stage 4, and 3 had progressed to stage 3 – that is, the disease stage had progressed in 5 of the 8 wrists while under nonsurgical treatment. At the endpoint, the median SLA was 68.5° (range 52°–84°), which indicated that 5 wrists had DISI. Radiographs from a representative case (case 4) at the initial visit and at the endpoint are shown in Figures 2 and 3. The median area reduction ratio of the scaphoid was 11% (range 4%–52%) on anteroposterior views and 4% (range −23% to 17%) on lateral-view radiographs, respectively.

Figure 1

The relationship between Herbert–Lanzetta classification and the follow-up periods, which depicted Herbert–Lanzetta classification on the Y-axis and duration of follow-up on the X-axis. Thin line (solid line: case 1, dotted line: case 2, dashed line: case 3, dash-dot line: case 4), thick line (solid line: case 5, dotted line: case 6, dashed line: case 7, dash-dot line: case 8).

Figure 2

Measurement of the area reduction ratio of the scaphoid on anteroposterior radiographs. The area of the scaphoid was traced and calculated at the time of their first visit (A) and at the endpoint (B). (A) Anteroposterior radiograph of the right wrist from a representative case (case 4) at the initial visit showing the increased density of the proximal pole scaphoid bone. The area of the scaphoid was traced and calculated using image analysis software (the area: 272 mm2). (B) Radiograph at the endpoint showing marked collapse of the proximal pole of the scaphoid bone without pathological fracture. The area of the scaphoid was traced and calculated using image analysis software (the area: 170 mm2). The area reduction ratio was 38%.

Figure 3

Measurement of the area reduction ratio of the scaphoid on lateral-view radiographs. The area of the scaphoid was traced and calculated at the time of their first visit (A) and at the endpoint (B). (A) Lateral radiographs of the right wrist from a representative case (case 4) at the initial visit. The area of the scaphoid was traced and calculated using image analysis software (the area: 200 mm2). (B) Radiograph at the endpoint showing marked collapse of the proximal pole of the scaphoid bone. The area of the scaphoid was traced and calculated using image analysis software (the area: 166 mm2). The area reduction ratio was 17%.

Clinical results

At the endpoint, compared with the findings in the contralateral wrist, the median wrist flexion–extension arc was 61% (range 50%–79%), and the median grip strength was 39%. The median MMWS was 17.5 points (range 10–25) – that is, poor in 6 patients (the flexion–extension arc and grip strength were not investigated in the other 2 patients).

Discussion

AVN of the scaphoid (Preiser disease) has been reported in only a few case series in the English-language literature.[ Also, AVN of the scaphoid and lunate (Preiser disease with concomitant Kienböck disease) is rare, having been reported in only 6 patients, including our 2.[ Although the etiology of AVN of the wrist is incompletely understood, most cases have involved steroid use[ and involved risk factors such as smoking, alcoholism, and/or infection.[ Moreover, undeveloped vascular networks and mechanical stress (eg, repetitive strain, overloading), which induce joint swelling and interruption of the intraosseous and extraosseous blood supply, could be predisposing factors.[ Although 2 of our patients had a history of repeated steroid injection of the contralateral shoulder or around the affected wrist, respectively, the other 6 did not have steroid use. However, 5 of the 8 had a smoking habit and a history of daily alcohol intake. Moreover, nearly all of our patients had a job that entailed repetitive mechanical stress on the wrist (Table 1). These factors may have contributed to the onset of Preiser disease in our patients.

Although radiographic staging guides treatment, early radiographic changes of AVN of the wrist are mostly undetectable, resulting in initial nonsurgical treatment. Previously, some researchers reported favorable outcomes of nonsurgical treatment for Preiser disease in a few cases.[ Others, however, reported that nonsurgical treatment – including immobilization, oral analgesics, corticosteroid injections, and electrical stimulation, among others – have not provided satisfactory outcomes.[ Moreover, the natural history of Preiser disease would presage unfavorable results, suggesting residual wrist pain and radiographic evidence of progressive deterioration. Although our patients were initially treated nonsurgically (ie, wrist immobilization with a splint or orthosis, rest) for various periods, it was ineffective.

Radiographically, the SLA was increased in most patients, and DISI was observed in 5 of our 8 patients. The area reduction ratio of the scaphoid was increased on anteroposterior radiographs, suggesting that almost all patients with AVN of the wrist exhibit radiographic progression and collapse of the scaphoid. Although the area reduction ratio of the scaphoid on lateral radiographs showed an increase in some patients, it could indicate that the shape of the scaphoid was collapsed and flattened. The scaphoid changes may initially occur as a cavitation due to AVN and result in flattening, collapse, and fragmentation caused by compression from the axial load from the radial portion of the articular surface of the distal radius. Moreover, all patients with AVN of the wrist exhibited persistent wrist pain. In addition, there was no obvious improvement of the wrist flexion–extension arc or grip strength compared with that of the contralateral wrist.

Based on our results, we concluded that nonsurgical treatment did not prevent deterioration of the radiographic findings and did not alleviate the wrist pain, ROM, grip strength, or MMWS. Because nonsurgical treatment had no positive effect on radiological or clinical improvement, and the scaphoid was at immediate risk of collapsing, prompt surgical intervention was indicated after arriving at the definitive diagnosis of AVN of the scaphoid. Nevertheless, there is still no consensus on the optimal surgical procedure for Preiser disease (with or without concomitant Kienböck disease) because of the disease's rarity.

Among various surgical procedures, 3 main protocols have been proposed for Preiser disease: revascularization surgery; salvage surgery; joint-leveling surgery. According to the Herbert and Lanzetta staging system, or depending on the arthritic location in the wrist described by Watson and Ballet,[ different procedures are indicated for different stages.[

Among these procedures, however, only revascularization surgery using a vascularized bone graft (VBG), mainly performed for early (stage 1–2) disease (Herbert–Lanzetta classification), could be a possible solution to simultaneously prevent scaphoid collapse and provide revascularization and augmentation of the bone strength of the scaphoid. Moran et al[ performed VBG from the distal radius in 8 patients with Preiser disease and reported good clinical outcomes for 2 patients with stage 1 to 2 disease. Also, as for the patient with Kienböck disease, several researchers reported that VBG provides an avascular lunate along with revascularization, increasing the bone strength of the lunate and possibly preventing its collapse.[ Because VBG allows preservation and survival of osteocytes,[ it provides accelerated graft consolidation in necrotic areas.[ Moreover, several studies have shown that VBGs provide more favorable mechanical and biological features than nonviable bone grafts.[ Thus, VBG might a good option for the avascular scaphoid to prevent carpal collapse in patients with early-stage disease.

To date, several authors reported the effectiveness of metaphyseal core decompression or radial osteotomy with or without radius shortening for Kienböck disease.[ Our previous study; however, showed that radial osteotomy for Preiser disease did not prevent scaphoid collapse.[ Thus, radial osteotomy might not be a solution for preventing collapse of the scaphoid. Further investigation regarding the effectiveness and outcomes of radial osteotomy for AVN of the carpal bone is required.

The major limitations of this study are the small sample size and its retrospective nature. Moreover, the study provides no comparison between treatment groups. Although our case series revealed that nonsurgical treatment for Preiser disease did not have a positive effect on the clinical and radiologic outcomes, further investigation regarding the effectiveness and outcomes of nonsurgical treatment for AVN of the carpal bone is needed.

Conclusion

Nonsurgical treatment for Preiser disease did not improve subjective or objective outcomes and did not prevent deterioration of radiographic findings.

Acknowledgment

We thank Nancy Schatken, BS, MT(ASCP), from Edanz Group (www.edanzediting.com/ac), for editing a draft of this manuscript.

Author contributions

Conceptualization: Yuji Tomori.

Data curation: Yuji Tomori.

Investigation: Yuji Tomori.

Methodology: Yuji Tomori.

Project administration: Yuji Tomori.

Supervision: Mitsuhiko Nanno, Shinro Takai.

Writing – original draft: Yuji Tomori.

Writing – review & editing: Yuji Tomori, Mitsuhiko Nanno.