Osteoid osteoma in the bones of the hand: a systematic literature review

Osteoid osteoma (OO), first reported by Jaffe 1935, is a benign bone tumor which is mostly present in the tibia, femur and other long bones, but may also occur at atypical locations [17, 30]. Its typical characteristic in imaging such as radiography, MRI (magnetic resonance imaging) and CT is a central nidus that contains high amounts of osteoid and is surrounded by sclerotic osseous tissue [10, 20]. The etiology of OO is unknown. The tumor is most commonly reported in patients under 30 years of age, with men more likely to be affected [18, 20]. The most typical symptom is nocturnal pain which is primarily sensitive to aspirin, but also other non-steroidal anti-inflammatory drugs (NSAIDs). This is most likely explained by the fact that the nidus inside the osteoid osteoma produces a high amount of prostaglandins [5].

OO in the bones of the hand are rare and, therefore, a diagnostic challenge. Most frequently they occur in the phalanges (59.2%) followed by carpal (30.1%) and metacarpal (10.7%) bones [25]. Many patients with OO have a long history of pain until diagnosis. Radiography and MRI can be less sensitive and lead to misdiagnosis, such as arthritis or cystic lesion [40, 43]. OO in the distal phalanges can also lead to deformity with hypertrophy of the nailbed [23, 36]. Due to the frequent proximity to joints and many other relevant structures in the hand as well as the small size of the bones, there may be difficulties in the treatment of OO in the hand. An overview of diagnostic and therapeutic options appears to be important to improve the treatment of the tumor.

The purpose of this systematic review was to provide a thorough and detailed assessment of reported cases of OO in the bones of the hand. The objectives of this article were to analyze patient age, symptoms, the time to diagnosis, the sensitivity of the used diagnostic modalities, the frequency of the effected bones and the conducted treatments.

This article provides a systematic overview and detailed analysis of epidemiology, symptoms, diagnostic and treatment of OO in the hand bones. In the literature there is a frequent occurrence of OO in the second and third decade [2, 5, 18]. The average age in the cohort studied was 26 years, with most patients aged between 20 and 29 years (53%). A frequent occurrence of OO in male patients could also be confirmed in our review [10, 34]. Together, these epidemiologic data are consistent with the literature, although this review shows that relatively more women may be affected and the mean age is slightly higher (female 38.5%, male 71.1%) [30]. When examining the most affected bones, there is an increased incidence in the proximal phalanges. As expected, the appearance of OO in the carpal bones was less frequent.

OO can be localized in the cancellous (i.e., trabecular) or cortical bone, with special localizations, such as subperiosteal or juxta-articular (i.e., intra-articular) [8, 20]. In the hand, a closer look at the localization within a bone is interesting, since in particular the carpal bones are small, covered with cartilage and a joint-near position of the OO is probable. The localization of the lesion has only been described in 37 of the reviewed cases. As described in the literature, intracortical OO was most common [20, 33]. Subperiosteal localization, rarely reported in the hand bones, is less common and can be a diagnostic challenge when occurring juxta-articular [11, 12, 37]. Juxta-articular localization of OO can lead to joint pain mimicking monoarthritis, which is why these cases are often misdiagnosed [31]. The nidus, which is surrounded by sclerotic bone, is also harder to identify in imaging caused by the complex radiological anatomy of the joint [14]. In addition, OO can be masked by a marrow edema [31]. Six cases of subperiosteal and three cases of juxta-articular localization were reviewed. We did not observe a frequent occurrence of pain radiation or mobility restrictions in the adjacent joints.

Local pain that responds to NSAID and increases at night is considered a typical symptom of OO [5] [16]. In our review, local pain was confirmed as a leading symptom in more than 95% of the cases. In most cases, there was additional pain at night and an improvement was achieved by NSAID. The combination of nocturnal pain sensitive to NSAIDs in combination with a typical lesion in imaging should be indicative of OO [31]. A symptom that has been described particularly frequently in OO in the proximal phalanges is local swelling. In some cases, the swelling occurred before the onset of pain [6, 38, 44]. In many of the reviewed cases OO in the distal phalanges led to deformity with hypertrophy of the nailbed [1, 23, 36, 43]. Levy et al. also reported a case of a 14-year-old female with OO in the distal phalanx of the thumb causing epiphysial growth arrest [23]. Limitations in the mobility of the surrounding joints were described less frequently. In some cases, the symptoms, such as pain and swelling, radiated into the surrounding joints [6, 19, 29]. Many patients with OO have a long history of pain until diagnosis [1, 29, 43]. With an average delay until diagnosis of 19 months, this statement can be confirmed in our review. These results are consistent with the experiences of patients with OO treated in our department of hand surgery. Most patients have been describing complaints for several months. In addition, some have been treated with suspected diagnoses, such as overuse or arthritis. In most cases, a nidus visible on CT scan led to the correct diagnosis.

CT scans have a high sensitivity in identifying OO which typically show a central nidus surrounded by sclerosis [24]. Of all the imaging modalities used in the reviewed cases the CT showed the highest sensitivity with 93.1%. Thin-section CT scans should be used when OO is suggested in the bones of the hand [4]. CT scans also allow visualization and precise localization of OO and are, therefore, helpful for preoperative planning [15]. Plain Radiography and MRI can be less sensitive and can lead to misdiagnosis, such as arthritis or cystic lesion [40, 43]. X-ray, which is often performed to diagnose hand complaints, had the lowest sensitivity at 64.4%. In many cases, the typical imaging characteristic of OO were revealed in X-rays well after the onset of the symptoms [6, 13, 43]. MRI is often used when patients have a long history of pain in the hand combined with swelling and tenderness. The sensitivity of MRI in the cohort studied was slightly higher at 81.6% than stated in the literature in localizations other than the hand [9, 18]. In MRI imaging, OO is described as a visible nidus surrounded by reactive marrow edema within the affected bone [21, 28]. Encircling the bone there can be an increase of soft tissue mass [21]. Scintigraphy was often performed as complementary imaging and had a high sensitivity of 90.5%. Scintigraphy using Technetium shows a diffuse increased uptake in the area of the OO [15]. Because of the symptoms typical for OO (pain, swelling, tenderness) an increased uptake can also be misdiagnosed as painful monoarthritis [21]. Accordingly, Scintigraphy has a status as a supplementary imaging to the CT or MRI to avoid misdiagnosis. SPECT has been used rarely and the case number is too small to make a valid recommendation.

Most of the OO cases in the bones of the hand are treated surgically with en bloc excision or open curettage [15, 21]. Importantly, surgical treatment offers the advantage of histological confirmation. Difficulties arise due to the small diameter of the bones and the proximity to important neurovascular structures. Bone defects can be filled with autogenous cancellous bone, which can be obtained from the iliac crest or the distal radius [15, 21]. Recurrences are rare. In OO in the distal phalanges, a correction of deformities may be performed simultaneously. In our department of hand surgery, we have had good experiences with the surgical removal of OO performing an open curettage. Larger defects were filled with cancellous bone from the distal radius.

In recent studies, percutaneous radiofrequency ablation (RFA) under computed tomography guidance is used as an alternative treatment for OO in the hand [26, 39]. It leads to good outcomes and a low morbidity [39]. When used for lesions in the hand and feet RFA must be used with caution as there is a risk of subsequent osteonecrosis or thermal damage to adjacent neurovascular structures [22, 30, 32]. When treating OO by RFA, it is important to reach inside the nidus to eliminate the lesion entirely [39]. Similar results are described in a study which used CT guided percutaneous laser photocoagulation to treat 15 cases of OO of the hands and feet [45]. Important are the placement of the optical fiber in the center of the nidus and the adaptation of the energy dose to the size of the nidus. Some studies used thermocoagulation to treat OO in the hand with a low morbidity after procedure but a high recurrence rate [42].

Overall, only little complications have been described after treatment, which may be due to the fact that OO are benign tumors with a small diameter [41]. From all complications, recurrence has been most commonly reported [32]. Most recurrences occurred after thermocoagulation (5 out of 12 cases) followed by open curettage (10 out of 96 cases) and en bloc resection (8 out of 100 cases). There were only a few cases of OO treated by laser photocoagulation or radiofrequency with no complications described in the follow-up examination.

Some limitations should be noted regarding this review. As with any systematic literature review, this study was directly limited by the quality of included literature. The majority of included source literature was case reports and case series with an evidence level of 4 or 5. This is also reflected in the results of the Quality assessment with a low average score. Only information that was explicitly described in the studies could be collected. Valid statements on selected questions such as the quality of the pain and the localization of OO in the bone could not be made due to the lack of data. As a limitation of the presented review, it should be noted that rare cases are more likely to be published, which could affect the distribution of the affected bones. Regarding the surgical therapy of OO, additional information on access routes, difficulties and post-treatment schemes would be useful. To determine the number of complications, especially of recurrences, a longer time to follow up in some of the case reports and case series would be necessary.

We reviewed 401 cases of OO in the bones of the hand. OO was most common in the proximal phalanges followed by the metacarpals and the distal phalanges. The frequency of typical symptoms such as pain, night pain and response to NSAIDs could be confirmed in the reviewed cases. Furthermore, if swelling and nail hypertrophy occurs, the diagnosis of OO in the phalanges should be considered. In some cases, motor limitations of the surrounding joints have been described. Together, the time to diagnosis is often prolonged in patients with OO in the bones of the hand. CT has the highest sensitivity and should be performed in case of suspicion of an OO. Most OO on the hand are surgically treated with excellent results. In recent years, studies with alternative therapies such as radiofrequency ablation, laser photocoagulation and thermocoagulation have also been shown to be effective.