Basic calcium phosphate deposition disease – a clinical and imaging analysis

Background. Basic calcium phosphate (BCP) deposition disease is a frequent musculoskeletal problem characterized by the intraarticular or periarticular deposition of carbonate substituted hydroxyapatite, octacalcium phosphate and tricalcium phosphate. BCP and calcium pyrophosphate (CPP) crystals play an important role in the pathogenesis of osteoarthritis. Objectives. The primary aim was to evaluate the clinical and ultrasonographic characteristics of patients with BCP. The secondary aim was to compare the radiographic scores of patients with BCP, calcium pyrophosphate dihydrate (CPPD) deposition disease and degenerative disease (DD). Material and methods. 50 patients with BCP deposition disease diagnosed by imaging or by identification of BCP in synovial fluid (SF) were included in the study. The second part of the study included 20 patients with BCP crystals in SF, 20 patients with CPP crystals in SF and 20 patients with degenerative changes on radiography and without crystals in SF. Clinical, ultrasonographic and radiographic data were recorded. Results. The shoulder joint (56%), followed by the knee joint (36%) and the acute clinical presentation (84%) were the most common findings. The fragmented ultrasound appearance (54%) was the most frequent, followed by the arch-shaped (36%). The localization of the deposits was found most often in the tendons (68%), mainly in the supraspinatus tendon (55.88%). Higher K/L scores were found in patients with BCP and CPP crystals than in the patients with DD. K/L score ≥3, defining osteoarthritis was associated with the presence of BCP and CPP crystals. Conclusions. The shoulder and the fragmented ultrasound pattern are the most common findings in patients with BCP. Higher K/L scores are found in patients with BCP and CPP crystal associated disease than in the degenerative disease patients.


INTRODUCTION
Basic calcium phosphate (BCP) deposition disease is defined as the deposition of BCP crystals intraarticular or periarticular. BCP crystals are containing carbonate substituted hydroxyapatite, octacalcium phosphate. From a clinical point of view, the deposits can be asymptomatic, or with an acute or chronic presentation. The shoulder is the most affected joint, mainly the rotator cuff, followed by the hips. Intraarticular deposition of crystals leads to a destructive arthropathy, known as Milwaukee shoulder (1)(2)(3)(4). The BCP crystals are formed when extracellular inorganic pyrophosphate (PPi) is low, mainly when inorganic phosphate (Pi)/PPi rapport is over 70 (5,6). The crystals have a complex effect on chondrocytes, macrophages and fibroblasts and also have the capacity to activate the nucleotide-binding domain, leucine-rich-containing family, pyrin domain-containing-3 (NLRP3) inflammasome. These mechanisms lead to inflammation, articular destruction, which can finally lead to osteoarthritis (3,(7)(8)(9)(10)(11). The gold standard for diagnosis is the identification of crystals in synovial fluid (SF), but due to the small dimensions of the crystals and tendance to form aggregates are difficult to be identified by light microscopy. Novel techniques are available but are expensive and complex (12). The calcifications are frequently discovered by radiological examination, having different sizes and different appearances (13,14). Ultrasound (US) is also a valuable method for the identification of BCP deposits (15)(16)(17).
This study aimed to describe clinical and imaging characteristics of patients with BCP. Our secondary aim was to compare the Kellgren-Lawrence score (K/L) score in patients with BCP, calcium pyrophosphate dihydrate (CPPD) deposition disease and degenerative disease (DD).

MATERIAL AND METHODS
In this transversal study were included patients with BCP crystals in SF or with evidence of BCP (radiographic or by US). Patients with gout or other inflammatory rheumatic diseases were excluded from the study. Data about age (years), sex, number of tenders (TJC) and swollen joints (SJC), erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP), the affected joints, clinical presentation as acute or chronic were recorded. US examination was performed using an Esaote My Lab 25 Gold with 5-10 and 10-18 MHz linear probes according to musculoskeletal guidelines. The protocol included searching for calcific deposits in symptomatic joints, identification of US pattern of calcification described for BCP (arch-shaped, fragmented, nodular, cystic), the affected joint, and the localization of deposits in the tendon, intraarticular, bursae, recess or other tissues (15)(16)(17)(18)(19).
BCP subgroup: A subgroup of 20 patients with BCP in SF was compared with 20 patients with CPPD and with 20 patients with DD. All the diseases were at the knee level.
CPPD group: Patients who fulfilled McCarty criteria for CPPD (imaging evidence of CPPD and crystals in SF) were included in the study (20).
DD group: Patients with degenerative changes on radiography and without crystals on SF analysis were consecutively included in the study.
Age, TJC, SJC, the visual analogue scale for pain (VAS), ESR (mm/h), CRP (mg/dl) were recorded for these three groups. All patients from the BCP subgroup, CPPD and DD group underwent radiographs in posteroanterior, weight-bearing semi-flexed view. All radiographs were graded according to K/L score (21). The score was interpreted at both knees, but also a medium K/L score, as a medium between the left and the right knee score was calculated. Osteoarthritis was defined as a K/L score ≥ 3. BCP and calcium pyrophosphate (CPP) crystals were identified in the SF according to the description of crystals (12,22).
All patients included in the study consented for participation and the approval of the University Ethics Committee was obtained.
Statistical analysis was performed using Microsoft Excel. We considered a level of p < 0.05 statistically significant.

RESULTS
The study included 50 patients with BCP deposition disease. The characteristics of the study population are presented in Table 1. The second, fragmented pattern was the most frequent in patients with BCP (Table 2). Regarding the location, the tendon was the most affected location and the supraspinatus the most common implicated ( Table 2). The fragmented appearence was more common in patients with acute presentation and monoarticular involvement (Figure 1).

BCP subgroup, CPPD and DD group evaluation
Comparing the three groups, significant differences were found between BCP, CPPD and DD regarding TJC, SJC, VAS and ESR. The results are described in detail in Table 3. Significant higher K/L scores were found in patients with crystal disease than in patients with DD at the right and left knee, but also when was calculated medium K/L score (Table 4). Between the BCP and CPPD group wasn't found any statistical significance (p = 0.89 for medium K/L score, p = 0.9 for right knee and p = 0.6 for the left knee.  Table 5.

DISCUSSION
The first part of the study examined the clinical and US characteristics of patients with BCP. Similar to other studies the shoulder was the most frequent joint involved and the clinical presentation was acute and monoarticular (3,4). The second most affected joint in our study was the knee, in contradiction with other studies in which predominant was the hip joint (2,4). The second, fragmented US pattern was the most frequent, probably because all patients were symptomatic and is known that the resorptive phase is associated with pain (15)(16)(17). In concordance with other studies, the most frequent localization of the deposits is in the supraspinatus tendon (3,4).
The second part of the study analyzed K/L score of three groups with BCP, CPPD and DD. BCP and CPP crystals are found in the SF of patients with osteoarthritis (23,24). Fuerst et al. found BCP crystals in all patients undergoing joint replacement for severe osteoarthritis (24). In clinical practice, many patients remain underdiagnosed due to difficulty to identify crystal with usual techniques (12). Both of these calcium-containing crystals induces inflammation and joint damage through complex mechanisms on chondrocytes and other cells. Also, CPP and BCP crystals have the property of activating the inflammasome NLRP3 (3,(8)(9)(10)25,26). For BCP are described other pathways too: increase the production of matrix metalloproteinases, tumour necrosis factor-alpha, interleukin 6 and Prostaglandin E2 (3,(7)(8)(9)(10)26). In the phatogenesis of osteoarthitris are implicated many factors. One of the most controversial is inflammation, mainly because osteoarthritis is not a disease characterized by inflammation (11,(27)(28)(29).
From a clinical point of view, the Milwaukee shoulder can sustain that inflammatory pathway.
Our results found higher values of TJC, SJC and VAS in the crystal associated groups than in the DD group. Similar results were found in a previous study in which was compared CPPD with osteoarthritis patients and in another study regarding pain, but not with arthritis (30,31). In our study significant higher K/L scores were found in patients with BCP and CPPD than in patients with DD, but not between the crystal associated groups. Also, K/L score≥ 3, which defines osteoarthritis was associated with the presence of BCP and CPPD. Higher radiographic scores were correlated to the presence of BCP crystals in a study by Derfus et al. too (32). In a recent study comparing the presence of CPP and BCP crystals in patients with osteoarthritis according to K/L score, there was not found any statistical significance between groups (33). Also, some studies found an association between osteoarthritis and chondrocalcinosis (34,35). The small number of patients and the absence of a structural phenotype analysis of osteoarthritis is the main limitation of the study.

CONCLUSION
The shoulder and the knee joint are the most affected joints. The fragmented US pattern is the most common and associated with an acute clinical presentation. Higher radiographic scores are associated with BCP and CPP crystal disease.