Drs. Gutierrez and Di Geso contributed equally to this work.
Special Theme Article: Clinical Imaging and the Rheumatic Diseases
Ultrasound Detection of Cartilage Calcification at Knee Level in Calcium Pyrophosphate Deposition Disease
Article first published online: 24 DEC 2013
Copyright © 2014 by the American College of Rheumatology
Arthritis Care & Research
Volume 66, Issue 1, pages 69–73, January 2014
How to Cite
Gutierrez, M., Di Geso, L., Salaffi, F., Carotti, M., Girolimetti, R., De Angelis, R., Filippucci, E. and Grassi, W. (2014), Ultrasound Detection of Cartilage Calcification at Knee Level in Calcium Pyrophosphate Deposition Disease. Arthritis Care Res, 66: 69–73. doi: 10.1002/acr.22190
- Issue published online: 24 DEC 2013
- Article first published online: 24 DEC 2013
- Accepted manuscript online: 7 OCT 2013 12:28PM EST
- Manuscript Accepted: 24 SEP 2013
- Manuscript Received: 10 MAR 2013
To determine the sensitivity, specificity, and accuracy of ultrasound (US) in the detection of cartilage calcification at knee level in patients with calcium pyrophosphate deposition disease (CPDD) and to assess the interobserver reliability.
Seventy-four CPDD patients and 83 controls with other chronic arthritis were included. All patients underwent a clinical examination, synovial fluid analysis, and radiographic assessment of the knee. US examinations were performed in order to detect hyperechoic spots within the hyaline cartilage layer and hyperechoic areas within the meniscal fibrocartilage. Twenty patients were assessed by 2 operators in order to calculate the interobserver reliability.
A total of 314 knees in 157 patients (74 with CPDD, 19 with rheumatoid arthritis, 17 with spondyloarthritis, 32 with osteoarthritis, and 15 with gout) were assessed. In the 74 patients with CPDD, hyaline cartilage spots were detected by US in at least 1 knee in 44 patients (59.5%), whereas radiography detected hyaline cartilage spots in 34 patients (45.9%) (P < 0.001). Meniscal fibrocartilage calcifications were detected by US in 67 of the 74 CPDD patients (90.5%), whereas conventional radiography detected calcifications in 62 patients (83.7%) (P = 0.011). The criterion validity expressed as percentage of sensitivity, specificity, and accuracy of US in the detection of articular cartilage calcification was high. Both kappa values and overall agreement percentages showed moderate to excellent agreement.
US is an accurate and reliable imaging technique in the detection of articular cartilage calcification at knee level in patients with CPDD.
Calcium pyrophosphate deposition disease (CPDD) is a disorder characterized by intraarticular and/or periarticular deposition of calcium pyrophosphate (CPP) crystals ([1-3]). The knee is an elective anatomic site for CPP crystal deposition with involvement of both hyaline and meniscal fibrocartilage ([1, 4-6]).
To date, conventional radiography is considered the routine imaging modality to detect cartilage calcification despite its relatively low sensitivity ([1-4, 7, 8]). Ultrasound (US) is gradually gaining interest among rheumatologists for the assessment of CPDD because it has demonstrated both high specificity and sensitivity in the detection of pathologic findings indicative of CPP crystal deposits ([9-18]). Its usefulness has also been emphasized in the most recent European League Against Rheumatism (EULAR) evidence-based recommendations for CPDD (). Despite these favorable characteristics, its validity issues largely remain to be investigated.
The main aim of the present study was to determine the sensitivity, the specificity, and the accuracy of US in the detection of articular cartilage calcification at knee level using microscopic analysis of synovial fluid as the gold standard. The secondary aim was to assess the reliability (interobserver variability) in the US detection of articular cartilage calcification.
Box 1. Significance & Innovations
- Ultrasound (US) is highly useful in the detection of calcium pyrophosphate (CPP) crystals at knee level, playing a key role in the diagnosis of calcium pyrophosphate deposition disease (CPDD).
- US can detect CPP crystal deposits at both the hyaline cartilage and fibrocartilage level despite negative findings on radiographs.
- US is an accurate and reliable imaging technique to detect cartilage calcification at knee level in patients with CPDD.
PATIENTS AND METHODS
Seventy-four patients with a crystal-proven diagnosis of CPDD by compensated polarized light microscopy (diagnosis according to McCarty criteria) () were consecutively included in the present study.
Eighty-three patients (19 with rheumatoid arthritis, 17 with seronegative spondyloarthritis, 32 with osteoarthritis, and 15 with gout) not having CPP crystals in the synovial fluid analysis were recruited consecutively as disease controls. For this aim the historical synovial fluid analysis recorded in our central database was considered.
CPDD patients and controls were eligible if the following additional criteria were present: pain in at least 1 knee and availability of conventional radiography of the knees in both anteroposterior and lateral views performed in the previous 3 months. Exclusion criteria were being age <18 years and having a history of severe trauma or knee surgery.
All patients were selected from those attending the outpatient and inpatient clinics of the Rheumatology Department of the Università Politecnica delle Marche.
The study was conducted according to the Declaration of Helsinki and local regulations. The institutional ethics committee approved the study and informed consent was obtained from all patients.
Both CPDD patients and disease controls underwent a clinical examination by an expert rheumatologist (RG) who recorded the presence/absence of spontaneous pain, tenderness (evocated by palpation and/or active and passive mobilization), and swelling at both knees.
In a second assessment the radiographic images were evaluated by a radiologist (MC) experienced in evaluating musculoskeletal system disorders and who was blinded to both clinical and US findings. The presence of cartilage calcification was reported and the radiologist registered the site of the calcification (hyaline cartilage or menisci). The purpose of the radiologist's evaluation was to confirm the rheumatologist's opinion.
All US examinations were performed by 2 rheumatologist sonographers (MG and LDG) with different experience in US (>8 years' experience for one rheumatologist and 3 years' experience for the other). Patients were asked not to talk about their clinical condition with the US examiners, who were blinded to clinical, laboratory, and radiographic data. US examinations were performed independently by the second sonographer, who was blinded with respect to findings of the first sonographer ([10, 12, 17]).
Before the study, the investigators reached a consensus on which US scanning technique to adopt and which pathologic findings to report in advance. Moreover, 20 subjects (10 with CPDD and 10 disease controls) were independently assessed in order to assess the interobserver reliability.
US examinations were performed using a Logiq 9 (General Electric Medical Systems) equipped with a multifrequency linear probe (9 MHz) or a MyLab 70 XVG (Esaote) equipped with a broadband multifrequency linear transducer (5–13 MHz).
The US scanning technique was performed according to the EULAR guidelines for musculoskeletal US in rheumatology for knee assessment ([12, 16]), including suprapatellar views with the knee in maximal flexion and posterior views with the knee extended, as previously reported (). To obtain the maximal exposure of the femoral cartilage surface, all patients were asked to perform the maximal degree of knee flexion.
In order to assess a wider cartilage surface, additional scans such as continuous sequences of parapatellar views were carried out with the knee in maximal flexion. Both medial and lateral aspects of the femoral cartilage not covered by the patella were explored on both longitudinal and transverse views. A dynamic examination during both compression with the probe and flexion–extension of the knee was carried out to identify the superficial margin of the hyaline cartilage. Lateral and medial longitudinal views during flexion–extension of the knee were the scans used to investigate the presence of meniscal calcification.
Parameters of gray-scale gain were initially set in order to obtain the maximal contrast between the different tissues under examination, and then were reduced to the lowest level in order to enhance the identification of the CPP crystals, allowing the visualization of only the structures presenting the same reflectivity of the bone.
US image interpretation
The presence of US pathologic findings indicative for cartilage calcification was systematically investigated for both knees. At hyaline cartilage level, the morphostructural changes used to detect the presence of CPP crystal deposits were the hyperechoic spots within the cartilage layer. The US identification of meniscal calcification depended on the detection of hyperechoic areas within the meniscal fibrocartilage showing similar echogenicity of the bony cortex even at very low level of gain. Figure 1 shows representative US images of CPP crystal deposits within both the hyaline and the meniscal cartilage.
We expressed standard descriptive statistics, including proportions for US findings and means and SDs for demographic data. The difference in terms of presence/absence of calcified cartilage by radiography or by US was calculated using Fisher's exact test for categorical variables. The level of significance was a P value less than 0.05.
Interobserver agreement was calculated by overall agreement (percentage of observed exact agreement) and unweighted kappa (κ) statistics for dichotomous scoring (e.g., presence/absence of calcified cartilage). All statistical analyses were performed using MedCalc, version 10.0 for Windows XP (Microsoft).
A total of 314 knees in 157 patients (74 with CPDD, 19 with rheumatoid arthritis, 17 with spondyloarthritis, 32 with osteoarthritis, and 15 with gout) were assessed. The mean ± SD (range) for age in CPDD patients and disease controls was 63 ± 9.2 years (39–79 years) and 64.5 ± 9.3 years (28–85 years), respectively.
In the 74 patients with CPDD, hyaline cartilage spots were detected by US in at least 1 knee in 44 patients (59.5%), whereas radiography detected hyaline cartilage spots in 34 patients (45.9%) (P < 0.001). Meniscal fibrocartilage calcifications were detected by US in 67 of the 74 CPDD patients (90.5%), whereas conventional radiography detected calcifications in 62 patients (83.7%) (P = 0.011). Tables 1 and 2 show the relationship between US and radiography findings indicative of hyaline cartilage and meniscal fibrocartilage calcifications.
|US findings (hyperechoic spots in cartilage layer)|
|Total (P < 0.001)||34 (45.9)||40 (54.1)||74|
|Total (P = 0.011)||62 (83.8)||12 (16.2)||74|
In the control group, 4 patients (1 with rheumatoid arthritis, 2 with osteoarthritis, and 1 with spondyloarthritis) showed US hyperechoic spots in hyaline femoral condyle and meniscal calcifications (Tables 3 and 4). After the US findings, a new synovial fluid analysis was performed in these 4 patients, which confirmed the presence of CPP crystals, revealing the coexistence of 2 pathologic conditions. Moreover, 2 patients showed calcifications on radiographs according to the radiologist's assessment (previously considered negative by the rheumatologist's assessment).
|Total (P = 0.036)||1 (1.2)||82 (98.8)||83|
|Total (P = 1.000)||1 (1.2)||82 (98.8)||83|
Table 5 shows the criterion validity expressed as percentages of sensitivity, specificity, and accuracy of US in the detection of articular cartilage calcifications and the values using microscopic identification of CPP crystals as the gold standard.
|Criterion validity||Microscopic identification of CPP crystals|
|Hyaline cartilage||Meniscal fibrocartilage|
In the 74 CPDD patients, hyperechoic spots within the femoral cartilage layer were more frequently found using the following scans: 45 (60.8%) for parapatellar, 20 (27%) for suprapatellar, and 9 (12.1%) for posterior views.
Exact agreement between the 2 sonographers was obtained in 91.4% of the scans with regard to hyperechoic spots within the hyaline cartilage and in 88.5% with regard to meniscal calcifications. Kappa values were 0.72 and 0.68, respectively.
This study provides evidence about the accuracy of US in the detection of cartilage calcifications at knee level in patients with CPDD. Although the use of US in the assessment of CPDD patients is still not an integral part of routine clinical practice, its role has acquired a growing interest in the last few years. As mentioned before, recently the EULAR recommendations have included US as a valid alternative imaging technique in the diagnostic approach to CPDD ().
Our results demonstrate a high specificity and good sensitivity, with a positive predictive value and a negative predictive value of US in identifying CPP calcifications at knee level. To date, only a few studies have been carried out in this topic, especially at the knee level. The study by Foldes showed a sensitivity of 89% and a specificity of 91% in the detection of calcification in femoral hyaline cartilage of the knee (). Similar results were successively obtained by Frediani et al () and more recently by Grassi et al (sensitivity 68.7%, specificity 97.6%, and accuracy 87.1%) () in the detection of hyaline cartilage calcifications. The present investigation confirms these data, but does so in a larger cohort of patients also using microscopic analysis and radiography, which are the gold standard and the most widely used imaging modality for the diagnosis of chondrocalcinosis.
Three of the 84 patients in the disease control group were sonographically positive for intracartilaginous hyperechoic spots, and 1 patient showed calcifications at the meniscal level. All of these patients presented with CPP crystals at the successive analysis of synovial fluid, revealing a coexistence of 2 pathologic conditions. This result is particularly relevant because it underlines the fact that conventional radiography sometimes could underestimate CPDD if concomitant with other disorders.
Another interesting finding of our study is the relatively high number of patients with the presence of cartilage calcifications at both the hyaline cartilage and the fibrocartilage level despite having radiographs with negative findings. There are several possible explanations for the relatively low sensitivity of radiography in the detection of cartilage calcifications. First, the bidimensional representation of the radiograph limits the whole assessment of the cartilage due to superimposition of the bone. Second, appropriate views for the visualization of femoral hyaline cartilage, such as lateral and tangential, are not carried out routinely. Third, pathologic concomitant conditions, such as severe knee osteoarthritis, that frequently present in the elderly and in patients with CPDD may impair the correct visualization of the cartilage because of the relevant joint space narrowing. Finally, other possible explanations for the discrepancy between sonographic and radiographic findings may be related to the size and density of CPP aggregates. The higher spatial resolution of US with respect to the radiograph can allow for the visualization even of minimal depositions. Moreover, conventional radiography may miss low density CPP aggregates, which are visible as hyperechoic spots using US ([20, 21]). These data suggest the use of US in patients with strong clinical suspicion of CPDD, in spite of negative radiographic findings, as an alternative or complementary tool. Furthermore, the safety of US for patients, the easy interpretation of the sonographic findings suggesting CPP deposits (even for an operator with limited experience in musculoskeletal US), and the possibility that US can be easily performed during a rheumatology examination may place it as the “ideal” tool in the diagnostic approach of CPDD.
A limitation of our study is the controls were not matched by possible confounder variables (i.e., age and sex), which could better support these data.
In conclusion, our study demonstrates that US is an accurate and reliable imaging technique in the detection of cartilage calcification at knee level in patients with CPDD.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Gutierrez had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Gutierrez, Di Geso, Salaffi, De Angelis, Filippucci, Grassi.
Acquisition of data. Gutierrez, Di Geso, Salaffi, Carotti, Girolimetti, Filippucci.
Analysis and interpretation of data. Gutierrez, Salaffi, Grassi.
- 3Calcium pyrophosphate dihydrate crystal deposition disease. In: Schumacher HR Jr, editor. Primer on the rheumatic diseases. 10th ed. Atlanta: Arthritis Foundation; 1993. p. 219–22..
- 17Tips and tricks to recognize microcrystalline arthritis. Rheumatology (Oxford)2012; Suppl 7:vii18–21., , .