SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Objective

Gaenslen's test (GT) positivity is characterized by tenderness upon lateral compression (squeezing) of the metacarpophalangeal (MCP) or metatarsophalangeal (MTP) joints. We aimed to assess the factors related to a positive GT and to explore differences in disease activity between GT-positive and -negative patients.

Methods

The GT was performed routinely in outpatients with rheumatoid arthritis (RA). In total, 229 patients had a positive GT (GT-positive group). To this group, we matched a GT-negative group (n = 222) comparable in sex, age, disease duration, and rheumatoid factor (RF) positivity. Disease activity scores, swollen and tender joint counts, patient pain and global assessments, evaluator's global assessment, morning stiffness, disability assessments, acute phase reactants, RF, and anti–citrullinated protein antibody levels were assessed and used for correlation analyses. A comparison between the GT-positive and -negative groups was performed and a regression model was calculated.

Results

All clinical variables, including disease activity scores, joint counts, and disability assessments, were significantly higher in the GT-positive group than the GT-negative group. Also, there were significantly fewer patients in remission or with low disease activity in the GT-positive group. GT positivity correlated most strongly (r > 0.5) with joint tenderness (32- and 28-joint counts). Tender proximal interphalangeal, MCP, and MTP joints appeared mostly responsible for a positive GT, but more than 1 MCP/MTP joint had to be tender to explain a positive GT.

Conclusion

The GT is related primarily to tenderness in MCP and MTP joints and signifies a moderate to high disease activity among patients with RA.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Rheumatoid arthritis (RA) is the most common chronic inflammatory joint disease with a prevalence of ∼1% worldwide (1). It is characterized by joint swelling and stiffness. The most frequently involved joints are those of the hands and feet, in particular the metacarpophalangeal (MCP), metatarsophalangeal (MTP), and proximal interphalangeal (PIP) joints. Differential diagnosis is not always easy, and the first step frequently relates to differentiation from noninflammatory joint diseases such as osteoarthritis. However, an early diagnosis mostly depends on early referral of patients with joint symptoms to the specialist; consequently, recommendations for referral of such patients are needed. Indeed, a proposal for criteria for early referral was published nearly a decade ago (2). Its mainstay was a squeeze test of the MCP and/or MTP joints. Interestingly, the squeeze test had previously been found to be an important factor for distinguishing early RA from other inflammatory and noninflammatory conditions (3), and in some new prediction models, it has been found to be useful as a diagnostic and prognostic test (4). This test was first described by the American surgeon Frederick Julius Gaenslen, who lived from 1877 to 1937 in Milwaukee and, therefore, the test is referred to as Gaenslen's test (GT).

Especially for primary care physicians, the GT is potentially a simple test for the assessment of early RA, although there are controversial opinions regarding if the GT is suitable for nonspecialists (2, 5, 6). A study has shown that the GT helps to differentiate between RA and other diseases like osteoarthrosis or fibromyalgia (7).

However, it is currently unknown which factors explain a positive GT. Moreover, before the GT can be recommended for use in early or established RA, data about its clinical utility, which are currently not available, are needed. It is also unknown if there is a relationship between the GT and disease activity. In order to investigate such a potential relationship, we performed analyses to determine 1) which factors mostly influence the GT in patients with RA, 2) if a special region of affected joints influences the GT, and 3) if there is a difference in disease activity between patients with a positive GT when compared with those with a negative GT.

Significance & Innovations

  • It is feasible to perform the Gaenslen's test (GT) routinely in a busy clinical setting.

  • The GT is related primarily to tenderness in the metacarpophalangeal and metatarsophalangeal joints.

  • The GT signifies a moderate to high disease activity among patients with rheumatoid arthritis.

Patients and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Patients.

Patients from our rheumatology outpatient clinic who had been diagnosed with RA according to the American College of Rheumatology 1987 criteria (5) and had a positive GT (GT-positive group; n = 229) participated in this study. To this group, we matched a GT-negative group (n = 222) comparable in sex, age, disease duration, and rheumatoid factor (RF) positivity. Exclusion criteria for the GT-negative group were any GT-positive assessment during the period of observation. The GT was performed routinely in outpatients with RA. The GT is a squeeze test of MCP and/or MTP joints and constitutes a purely dichotomous measure, where positive indicates that the patient describes tenderness and negative indicates no tenderness.

Patients seen in the clinic underwent evaluations of swollen and tender joints using 28-joint counts (SJC28 and TJC28, respectively) and 32-joint counts (SJC32 and TJC32, respectively; the 32-joint count is the 28-joint count plus ankles and MTP joints as single joints) (6), patient's pain (PPA) and global (PGA) assessments, evaluator's global assessment (EGA), the Health Assessment Questionnaire (HAQ) (7, 8), and morning stiffness (9), as well as calculations of the Simplified Disease Activity Index (SDAI) and Clinical Disease Activity Index (CDAI) (10, 11) at every visit according to established cut points for remission, low disease activity, moderate disease activity, or high disease activity (12). In addition, laboratory variables, such as acute-phase reactants (C-reactive protein [CRP] level and erythrocyte sedimentation rate [ESR]) and autoantibody testing (RF and anti–citrullinated protein antibodies [ACPA]), were assessed. The Ethical Committee of the Medical University of Vienna approved this study.

Statistical analyses.

To evaluate which variables were normally distributed, the Kolmogorov-Smirnov test was used. Except for age, none of the variables were normally distributed. Results are therefore expressed as medians and interquartile ranges. Differences between the groups were calculated using the Mann-Whitney test. To test for correlations between the GT and other variables, a univariate analysis was performed by Spearman's rank correlation. P values were corrected using the Bonferroni method. Subsequently, a regression model was calculated, entering all possibly explanatory variables. If variables depicted similar information and were significantly correlated with each other, only one of those variables was selected for calculating the model. P values less than 0.05 were considered statistically significant. SPSS statistical software, version 17.0, was used for the statistical analyses.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Demographics and baseline characteristics.

Table 1 depicts the demographic and clinical baseline characteristics of the 229 GT-positive patients and 222 GT-negative patients. As can be seen, several disease activity measures differed significantly between the groups. When we assessed disease activity states using established cut points (12), significantly more GT-negative patients were in remission and had low disease activity than GT-positive patients, whether using SDAI or CDAI (P < 0.001). Using SDAI, we found 39.7% of patients in the GT-positive group and 75.2% of patients in the GT-negative group were in remission or had low disease activity; when using CDAI for calculating disease activity instead, the proportions were 40.2% and 77.4% for the GT-positive and -negative groups, respectively.

Table 1. Baseline characteristics in the GT-positive and GT-negative groups*
 GT-positive group (n = 229)GT-negative group (n = 222)
  • *

    Values are the median (interquartile range) unless otherwise indicated. GT = Gaenslen's test; RF = rheumatoid factor; ACPA = anti–citrullinated protein antibodies; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate; PPA = patient's pain assessment; PGA = patient's global assessment; EGA = evaluator's global assessment; SJC28 = swollen joint count in 28 joints; SJC32 = SJC in 32 joints; MCP = metacarpophalangeal; PIP = proximal interphalangeal; TJC28 = tender joint count in 28 joints; TJC32 = TJC in 32 joints; HAQ = Health Assessment Questionnaire; SDAI = Simplified Disease Activity Index; CDAI = Clinical Disease Activity Index.

  • Significant difference (P < 0.05).

Age, mean ± SD years57.8 ± 12.558.4 ± 12.5
Disease duration, mean ± SD years11.3 ± 8.811.0 ± 7.3
Female, no. (%)195 (85.2)193 (86.9)
RF40.5 (12–168.8)41.0 (12–146.5)
ACPA73.5 (2–340)72 (1.7–340)
CRP level0.44 (0.19–1.4)0.42 (0.1–0.9)
ESR25 (14–54.8)21 (12.5–44.5)
PPA45 (26–60)23 (5–42)
PGA48 (27–62)24 (5–45)
EGA11 (5–22)4 (0–11)
SJC281 (0.5–3)1 (0–2)
SJC322 (1–4)1 (0–2)
Swollen MCP joints, mean ± SD1.0 ± 1.50.7 ± 1.3
Swollen PIP joints, mean ± SD0.9 ± 1.50.5 ± 1.0
TJC284 (1–8)0 (0–1)
TJC324 (2–10)0 (0–1)
Tender MCP joints, mean ± SD2.0 ± 2.70.4 ± 1.1
Tender PIP joints, mean ± SD1.6 ± 2.50.3 ± 1.1
HAQ1.1 (0.6–1.8)0.5 (0.03–1)
SDAI13.2 (7.4–21.5)5.7 (2.7–9.4)
CDAI12 (6.6–19.3)5 (1.7–9.1)
Morning stiffness, mean ± SD43.3 ± 67.410.8 ± 27.0
Glucocorticoids, no. (%)106 (46.3)43 (19.4)
Glucocorticoid dose0 (0–5)0 (0–0)

Correlations between the GT and clinical variables.

The correlation between the GT and other clinical, laboratory, and functional variables was calculated by Spearman's correlation (Table 2). The GT was significantly (P < 0.001) correlated with the TJC28 and TJC32 (and consequently also with the SDAI and CDAI); the correlation between the GT and the PPA, PGA, EGA, SJC28, SJC32, HAQ, and morning stiffness, as well as the region of affected tender joints revealed lower coefficients, although all correlations were statistically significant (P < 0.001). Interestingly, there was no correlation between the GT and laboratory variables such as CRP level, ESR, RF, and ACPA. Correlations were also separately calculated for the GT of the hands and feet with similar results, whereby the GT of the feet was best correlated with tender MTP joints (Table 2).

Table 2. Correlation coefficients between GT, hand GT, as well as foot GT and different variables*
 GTHand GTFoot GT
  • *

    All correlations were significant (P < 0.001) except where indicated. GT = Gaenslen's test; PPA = patient's pain assessment; PGA = patient's global assessment; EGA = evaluator's global assessment; SJC28 = swollen joint count in 28 joints; TJC28 = tender joint count in 28 joints; SJC32 = SJC in 32 joints; TJC32 = TJC in 32 joints; HAQ = Health Assessment Questionnaire; SDAI = Simplified Disease Activity Index; CDAI = Clinical Disease Activity Index; PIP = proximal interphalangeal; MCP = metacarpophalangeal; MTP = metatarsophalangeal.

  • Not significant.

  • Significant (P < 0.05).

PPA0.390.360.29
PGA0.380.350.27
EGA0.370.390.18
SJC280.240.310.04
TJC280.530.550.36
SJC320.270.310.08
TJC320.560.530.43
HAQ0.370.350.28
SDAI0.460.500.29
CDAI0.480.510.31
Morning stiffness0.370.390.24
Glucocorticoids0.290.270.11
Tender PIP joints0.310.420.24
Tender MCP joints0.410.350.18
Tender MTP joints0.300.190.52

Multivariate analysis.

When assessing correlations between the individual variables employed, the highest correlation was found between the PGA and PPA (rs = 0.90), the EGA and SJC28 (rs = 0.86), the EGA and SJC32 (rs = 0.88), and, expectedly, between the SJC28 and SJC32 (rs = 0.97), the TJC28 and TJC32 (rs = 0.97), the TJC28 and CDAI (rs = 0.82), the TJC32 and CDAI (rs = 0.82), and the SDAI and CDAI (rs = 0.98, P < 0.001 for all). Therefore, we selected 1 variable from each of the following categories for each model: tender joints (TJC28, TJC32, and tender MCP or MTP joints), swollen joints (SJC28, SJC32, and swollen MCP or MTP joints), patient-reported variables (HAQ and PPA), as well as disease activity variables (CDAI and SDAI). The 16 calculated models are shown in Table 3. Each model explained the variance of the GT up to a certain level (shown in the second row of Table 3). If this explained variance was higher, the variables entered into the model explained more of the variance of the GT. If the MCP or MTP joints were in the model, the explained variance was high and other variables were not significantly correlated. If other variables were entered into the model, the explained variance was not as good. Therefore, joint tenderness was the strongest predictor explaining the GT.

Table 3. Regression models 1–16*
 Model 1Model 2Model 3Model 4Model 5Model 6Model 7Model 8Model 9Model 10Model 11Model 12Model 13Model 14Model 15Model 16
  • *

    The dependent variable was the Gaenslen's test (GT). Each model explains the variance of GT up to a certain level (explained variance). The other variables entered into each model were selected based on correlations with GT and with each other. Asterisks in the table body indicate a significant contribution of the specific variable to the model; accordingly, dashes indicate no significant contribution and no entry means that this variable was not entered into the model. TJC32 = tender joint count in 32 joints; TJC28 = TJC in 28 joints; T_MCP = TJC of the metacarpophalangeal joints; T_PIP = TJC of the proximal interphalangeal joints; SJC32 = swollen joint count in 32 joints; SJC28 = SJC in 28 joints; S_MCP = SJC of the MCP joints; S_MTP = SJC of the metatarsophalangeal joints; HAQ = Health Assessment Questionnaire; PPA = patient's pain assessment; CDAI = Clinical Disease Activity Index.

Explained variance0.370.350.330.290.370.370.360.360.340.340.290.290.290.320.350.31
TJC32*   **        * 
TJC28 *    **        
T_MCP  *     **     *
T_PIP   *      ***   
SJC32            
SJC28           *
S_MCP  *         
S_MTP              
HAQ***********
PPA****** 
CDAI             *  

GT and tender joints.

The number of tender MCP and PIP joints in the GT-positive group were mean ± SD 2.0 ± 2.7 and 1.6 ± 2.5, respectively, whereas in the GT-negative group, mean ± SD 0.4 ± 1.1 tender MCP joints and 0.3 ± 1.1 tender PIP joints were affected (P < 0.001 for both).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

The GT is a frequently employed clinical procedure. The present analysis reveals the contextual framework of the GT. It could be shown that the TJCs were the strongest predictor of the GT, especially tender MCP, PIP, and MTP joints. Since performing the GT requires squeezing MCP and MTP joints, the correlation of the GT with tenderness of these joints is intuitive; the relationship of the GT with the presence of PIP joint involvement was less clear. However, an isolated affection of the PIP joints was not common (only in 6.6% of GT-positive patients and 7.2% of GT-negative patients) and, therefore, this appears to be an indirect relationship. Both the TJC28 as well as TJC32 predicted the GT. Interestingly, the SJCs did not correlate highly with the GT. Analyses of individual joints revealed that the GT depends on a polyarticular involvement of the MCP and MTP joints and therefore reflects polyarticular affection of the most frequently involved joints in RA.

The clinical significance of the GT goes beyond the mere explanation of the abnormalities associated with this squeeze test. The GT-positive patients had significantly higher disease activity by all clinical and functional means when compared with the GT-negative group. Interestingly, there was no significant correlation between the GT and laboratory variables such as ESR, CRP level, or the presence of autoantibodies. In line with the clinical findings, significantly fewer GT-positive patients were in remission or had low disease activity than GT-negative patients. Thus, we could show that the GT is associated with a higher disease activity. Therefore, the GT also reflects active disease, an important aspect in conjunction with patient referral, since patients with active disease will require more rapid attention by the expert and institution of appropriate therapy compared with patients with inactive disease.

A limitation of our study is the lack of standardization of the test. In our clinic, trained assessors perform the test in parallel with other clinical joint assessments. Clearly, having a device that employs a standardized pressure on the MCP and MTP joints might be helpful, although the effects of such a device could still vary among patients. However, the intention of our study was to define the abnormalities underlying the squeeze test, which until now has been frequently employed in clinical practice without good explanations of its background. As a further limitation, we have not assessed other patient populations to evaluate the differential diagnostic potential of the GT. However, the stimulus for the present study was not the differential diagnostic aspect, but our wish to understand the mechanisms underlying the GT. Furthermore, another limitation is that the GT was performed by 1 assessor only. However, our study focused on the potential clinical utility and not the reliability of the test.

While some authors have suggested that the GT is not widely used and not generalizable in routine practice (13), others have employed the test in routine care and even used it to predict RA (4). The reason for not using the GT in routine practice might be related to the lack of information on the abnormalities. These abnormalities have now been clarified in the present study.

In conclusion, the GT is primarily correlated with tenderness of the MCP and MTP joints. A positive compression test relates to the involvement of more than 1 joint of the MCP or MTP region and signifies moderate to high disease activity among patients with RA. While the differential diagnostic potential of the test will need to be examined in further studies, the data presented herein reveal that the GT is a useful reflection of the involvement of the joints most frequently affected in RA.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

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 published. Dr. Stamm 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. Wiesinger, Smolen, Aletaha, Stamm.

Acquisition of data. Wiesinger, Stamm.

Analysis and interpretation of data. Wiesinger, Smolen, Aletaha, Stamm.

REFERENCES

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES