Article Text

Concise report
Extent and distribution of CPP deposits in patients affected by calcium pyrophosphate dihydrate deposition disease: an ultrasonographic study
  1. Georgios Filippou1,
  2. Emilio Filippucci2,
  3. Marika Tardella2,
  4. Ilaria Bertoldi1,
  5. Marco Di Carlo2,
  6. Antonella Adinolfi1,
  7. Walter Grassi2,
  8. Bruno Frediani1
  1. 1Department of Clinical Medicine and Immunology, Rheumatology Section, University of Siena, Siena, Italy
  2. 2Clinica Reumatologica, Università Politecnica delle Marche in Jesi, Ancona, Italy
  1. Correspondence to Dr Georgios Filippou, Department of Clinical Medicine and Immunology, Rheumatology Section, Policlinico le Scotte, Viale Bracci 16, Siena 53100, Italy; g_filippou{at}


Objective To assess the extent of calcium pyrophosphate dihydrate (CPP) crystal deposition and the distribution of affected sites, using ultrasonography (US), in patients affected by CPP deposition disease (CPPD).

Patients and methods 42 consecutive patients affected by definite CPPD according to the McCarty criteria were enrolled in the study. All patients underwent an US examination of metacarpophalangeal joints of II, III, IV and V fingers of both hands, wrists and knees, Achilles’ tendons and plantar fascia looking for CPP deposits. A dichotomous score for presence/absence of CPP and a semiquantitative score for extent of deposits (0–3: 0, absent; 1, 1–2 spots; 2, more than two spots covering <50% of the structure; 3, deposits covering >50% of the structure) were assigned to each site examined. A site distribution score (total number of affected sites) was then calculated as well as an extent score equal to the sum of the extent scores of all sites.

Results The mean involvement in our patients was 4.7 sites (SD±1.7, range 2–8 sites). The knee was the most affected, site (41 of 42) followed by the wrist (at least one in 37 patients) the Achilles’ tendons (23 patients), plantar fascia (11 patients) and metacarpophalangeal joints (four patients). The highest mean values of the extent score were in the menisci, followed by the hyaline cartilage of the femoral condyles and the entheses.

Conclusions The deposition of CPP crystals involves at least two sites with a mean of four sites involved in most patients affected by CPPD and is therefore an oligoarticular or polyarticular disease.

  • Chondrocalcinosis
  • Ultrasonography
  • Arthritis

Statistics from


In the last decade, ultrasonography (US) has proven to be an excellent technique for detecting calcium pyrophosphate dihydrate (CPP) crystal deposits in joints and periarticular tissues.1–9 On behalf of EULAR, a group of experts recently published a set of recommendations for CPP deposition disease (CPPD), including terminology, diagnosis and management.10 The experts systematically reviewed the literature on the use of US to detect CPPD, and concluded that US detects CPP crystals in peripheral joints with excellent sensitivity and specificity, possibly better than those of conventional x-rays. This is the first time that US has been included among the diagnostic criteria for CPPD, even as a recommendation.

The aim of the present study was to assess the extent of CPP deposits and the number of affected sites using US in patients with this disease in order to define the most frequently involved sites.

Patients and methods

In this two-site study (Rheumatology Institute in of the University of Siena in Siena and Clinica Reumatologica of Università Politecnica delle Marche in Jesi, Ancona, Italy), we enrolled consecutive outpatients and inpatients with a definite diagnosis of CPPD according to the McCarthy criteria,11 having positive both x-rays and synovial fluid analysis for the presence of CPP crystals. All patients signed an informed consent for inclusion in this study. Local ethics committee approval was not necessary as all patients underwent US examination for diagnostic purposes according to our local protocols. All patients were negative on anamnesis for diseases that could lead to CPPD (eg, hyperparathyroidism). The cartilage of the metacarpal heads from the second to the fifth finger was scanned on the dorsal side of the hand with the fingers in maximum flexion. The cartilage of the femoral condyles was observed by anterior scans with the knee in maximum flexion and posterior scans with the leg extended. The triangular ligament of the carpus was studied with multiplanar scans, rotating the probe around the ulnar compartment of the wrist. The medial and lateral menisci of the knee were examined with multiplanar scans with the knee flexed at 30–45° and completely extended. US examination of the calcaneal entheses was conducted with the prone patient on the examination bed in order to study the Achilles tendon and the plantar fascia. The latest generation US systems (MyLab70XVG, Esaote Biomedica, Genoa, Italy) with linear 6–18 MHz probes were used in both centres. All US examinations were carried out by two experienced ultrasonographers.

To identify CPP crystals by US, we used the criteria proposed previously describing microcrystalline deposits as hyperechogenic bands and spots, generally without posterior shadow, in hyaline cartilage and fibrocartilage (see online supplementary image S1).4 ,5 Tendon calcifications appear as typical linear deposits, often distributed along the major axis of the tendon (figure 1). We preferred to screen only these sites as US aspect of CPPD has not been yet described accurately for other joints.

Figure 1

Calcium pyrophosphate dehydrate (CPP) crystal and other crystal ultrasonography (US) typical in the Achilles tendon. In longitudinal scan (A), CPP crustal deposit is seen as a linear hyperechogenic deposit positioned at the major axis of the tendon that does not create posterior shadowing (arrows). In transverse scan, the anisotropic artefact can been used to highlight CPP deposits (arrows). In (B), the tendon is correctly visualised and CPP deposits can hardly been seen (arrows). When the anisotropic artefact is present (C), tendon structure becomes hypoechogenic and CPP deposits are highlighted as they are not affected by anisotropy. This way more deposits, otherwise not seen, can be found (curved arrow). In panel D, the US aspect of urate monosodium deposits (tophi) in Achilles tendon can be observed (arrows) while in panel E the typical aspect of tendon calcifications (not due to CPP) with posterior shadowing (curved arrow) can be appreciated. Access the article online to view this figure in colour.

A dichotomous score for presence/absence of crystals and a semiquantitative score of the extent of deposits (0–3: 0, absent; 1, 1–2 spots; 2, more than two spots covering <50% of the volume of the structure; 3, deposits covering >50% of the volume of the structure explorable by US) were assigned to each site examined. A site distribution score (the sum of affected sites, 0–16) was then calculated for each patient, as well as an extension score, ranging from 0 to 78 equal to the sum of the scores of all sites, as the knee cartilage was divided into four compartments (anterior and posterior for the medial and lateral condyle) and menisci were counted as separate sites in order to improve accuracy and reproducibility of the scoring method as explained later in the Discussion section. Inter-rater agreement for the presence or absence of CPP between the two observers was calculated by Cohen's κ on five patients with confirmed CPPD, while inter-observer agreement for the scoring system has been calculated in a new set of five patients with CPPD and calculated by intraclass correlation. Logistic regression has been used for the correlation between clinical presentation and extent of the deposits. The SPPS software V.13 has been used for the statistical analysis.


The number of patients enrolled was 42 of whom 26 (62%) were female subjects. The mean age was 64 years (SD±17years). Two patients underwent monolateral knee joint replacement. Thus, a total of 82 knees were scanned. Moreover, in 16 out of 328 compartments of the femur condyles, the thickness of the cartilage was extremely reduced due to severe osteoarthritis and an accurate evaluation for the presence of crystals at these sites could not be performed.

The clinical presentation of CPPD in our cohort of patients, according to the EULAR task force,10 was asymptomatic, with or without osteoarthritis manifestations, in 16 patients (38%), acute CPP crystal arthritis in 10 (24%) and chronic CPP crystal arthritis in 16 (38%).

The highest mean values of the extent score were in the menisci followed by hyaline cartilage of the femoral condyles and entheses (table 1). The mean extent score was 18.57 (SD±10.2, range 3–47). The exact number of single sites with the various scores is illustrated in figure 2. The mean number of the sites involved by CPPD was 4.7 (SD±1.7) and the range was 2–8 affected sites. The knee was the most affected joint as 41 of 42 patients (97.6%) presented at least monolateral involvement, followed by the wrist, Achilles’ tendons, plantar fascia and metacarpophalangeal joints.

Table 1

Joint distribution and extent of CPP deposits in our cohort of patients

Figure 2

Number of sites with extent score of 0–3. The cartilage has been grouped in two compartments, anterior and posterior and metacarpophalangeal joints have been omitted for the very low number of joints involved (only seven of 336, six with score 1 and one with score 2). The number of meniscus and cartilage is lower than the sum of the patients as we also had two knee joint replacements and in some cases cartilage was too thin due to osteoarthritis to be evaluated.

No correlation was found between the clinical presentation of the disease and number of involved sites. Patients with the asymptomatic form had a mean site involvement of 4.81 sites (SD±1.682) and an extent score of 22.50 (±12.4), patients with chronic CPP arthropathy 4.81 (SD±2.007) and an extent score of 15.69 (±8), and patients with pseudogout had a mean of 4.5 (SD±1.269) sites with CPPD and an extent score of 16.9 (±8.2). Cohen's κ ranged from 1 (perfect agreement, external meniscus) to 0.588 (moderate agreement, hyaline cartilage) while intraclass correlation ranged from 0.911 (almost perfect agreement, menisci) to 0.581 (moderate agreement, triangular fibrocartilage).


CPPD is characterised by calcifications due to deposition of CPP crystals in joints and associated structures.11 ,12 CPPD seems to be a frequent disease especially in the elderly13 but the exact prevalence and incidence values are very difficult to determinate as the diagnosis of this disease is relatively difficult to obtain, in part because frequently it is asymptomatic and in part because it depends on the diagnostic tool we use.14 ,15 Furthermore, the natural history of the disease is not known, as there are no longitudinal studies. Thus, it is very important to investigate the impact of new imaging tools in monitoring the disease.

To the best of our knowledge this is the first study aiming at investigating the distribution and the extent of crystal deposits in patients affected by CPPD according to the criteria of McCarty.11 In our cohort, all patients had at least two sites involved with a mean site count of almost 5. This means that CPPD is a multifocal disease and could also indicate an underlying systemic process rather than only the presence of local factors for CPPD. In the group of patients with the asymptomatic form, the mean site count was similar (4.81) to the others meaning that probably the number of the joints involved does not influence the clinical presentation of the disease. The extent score of the deposits resulted higher in the patients with the asymptomatic form and lower in those with pseudogout and chronic arthritis. This could look strange, as someone could expect more deposits in the patients with arthritis. If we consider though the crystal shedding theory11 as responsible for the arthritis (either acute or chronic), we could hypothesise that patients with arthritis ‘shed’ more crystals than patients with the asymptomatic form in their joints and this could explain the major number of the deposits in the latter.

The greatest extent of deposition was found in menisci, followed by the triangular fibrocartilage of the wrists. Hyaline cartilage was more involved in the knees (femoral condyles) than in the hands, where CPP deposits were only found in four patients with almost exclusive involvement of the second metacarpal head (the right third and fourth heads were also involved in only one patient). The predisposition of fibrocartilage, with respect to hyaline cartilage, for microcrystalline deposits may provide new insights into the pathogenesis of crystal formation in tissues.

The scoring system adopted in the present study could be useful for monitoring the disease over time. In order to obtain a more accurate and reproducible scoring system we decided to divide the knee hyaline cartilage into four compartments, anterior and posterior of the medial and lateral condyle. In this way, it is easier for the observer to assign an extension score as the volume to explore is smaller than the whole cartilage of the knee. We also decided to assign the extent score at the explored volume of the structure under examination and not at a single image as the distribution of CPP crystals was not expected to be homogeneous (see online supplementary image S2). The sonographer had to score keeping in mind the whole structure and not only the scanning plan showing major involvement. Such an approach makes this scoring system particularly useful for the longitudinal assessment of patients by CPPD, avoiding the ‘ceiling effect’ of the evaluation on a single image, which would probably lead to a score of 3 in most of the cases. The progression of the US scores in relation to the clinical manifestations could provide new evidence on the natural history of the disease and treatment efficacy.

This study has some limitations; first, the inter-observer variability in evaluation of the presence/absence of deposits and in the estimation of their extent and distribution. Second, no controls were enrolled and sonographers were not blind to patients’ diagnoses. Finally, even if we had a percentage of asymptomatic cases in our cohort (mainly patients with OA of the knee and without inflammatory pain in the past) there could be a selection bias as all patients were hospital-referred. A population-based study could be more appropriate but it is also very difficult to perform.

In conclusion, the results of the present study indicate that the CPP crystal deposition involves more than one site in all CPPD patients and their extent is higher in patients with the asymptomatic form. An US scan of the knees could be a useful indicator of the disease as almost 98% of all patients have at least monolateral involvement. Also, at meniscal level, CPP identification should be easier as the extent of the deposits is larger than in other sites. We believe that the use of US for assessing the extent and the distribution of the CPP crystal deposits could help in characterising and monitoring patients with CPPD and requires further investigation.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Handling editor Tore K Kvien

  • Contributors All authors included on this paper fulfil the criteria of authorship. In addition, no one else who fulfils the criteria has not been included as an author.

  • Competing interests None.

  • Ethics approval Ultrasonographic examinations are routinely carried out in our departments in patients affected by CPPD.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.