Pathological features of rheumatoid synovitis in the context of therapeutic interventions and clinical settings

(1)

therapeutic interventions and clinical settings

Oosterhout, M. van

Citation

Oosterhout, M. van. (2008, November 19). Pathological features of rheumatoid synovitis in the context of therapeutic interventions and clinical settings. Retrieved from

https://hdl.handle.net/1887/13337

Version: Corrected Publisher’s Version

License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden

Downloaded from: https://hdl.handle.net/1887/13337

(2)

Chapter 2

Superior effect of arthroscopic lavage compared to needle aspiration in the treatment of inflammatory arthritis of the knee

Arthroscopic lavage versus joint aspiration

M van Oosterhout, JK Sont, JM van Laar

Rheumatology 2003 Jan;42(1):102-7

(3)

Abstract

Objective

To compare the duration of symptom relief after arthroscopic lavage versus needle aspiration in gonarthritis.

Methods

A retrospective chart analysis was performed in fifty patients with non-septic inflammatory arthritis of the knee who underwent arthroscopic lavage because of relapsing or persisting arthritis after needle aspiration. The primary outcome measure was the time period until symptomatic recurrence of knee synovitis.

Results

Needle aspiration was associated with a 3.0 times greater risk of recurrence of arthritis compared to arthroscopic lavage within twelve months (p<0.001, 95% confidence interval 2.1-4.4). Patients with longer disease duration and a higher number of used DMARDs had a significantly lower risk of recurrence of arthritis compared to patients with shorter disease duration and a lower number of previous DMARDs (p=0.04 and 0.02 resp.). Corticosteroids augmented the effect of both interventions.

Conclusions

Our results indicate that arthroscopic lavage is an effective therapeutic modality in the treatment of inflammatory arthritis of the knee refractory to joint aspiration, especially in patients with longstanding disease.

(4)

Introduction

Different therapeutic approaches are currently employed in the management of persistent or relapsing arthritis of the knee joint. Joint aspiration either with or without corticosteroid injection is commonly used, while synovectomy by various means is usually reserved for therapy resistant cases. Since the description of the beneficial effects of joint injection with steroids [1, 2] a number of publications have elucidated factors that contribute to a longer effect of the injection: 1. post injection rest [3], 2. the choice of a less soluble steroidcompound [4], 3. evacuation of synovial fluid before steroid injection [5].

Several studies found a therapeutic effect of arthroscopic procedures involving a lavage or washout of inflamed knees [6-9]. The therapeutic effect of both joint aspiration and lavage is attributed to the removal of cartilage degradation products, pro-inflammatory cells and destructive enzymes [8-13] as well as to an improvement of the anti-inflammatory effect of steroids. Joint aspiration yields immediate pain relief through decompression of the joint cavity [14, 15]. Furthermore, dilution of the steroid is greatly diminished when residual synovial fluid is minimised possibly resulting in a better response [5, 10]. Arthroscopic or closed needle lavage of the arthritic knee has been advocated as a potentially more effective alternative to joint aspiration because lavage would enable more extensive removal of debris and inflammatory mediators (9,12,13). This assumption is based on the longer duration of the washout procedure and the larger diameter of the portal used in arthroscopy. An additive effect of joint lavage to steroid injection was described in a small number of rheumatoid arthritis patients using 60 to 120 mL of saline [16]. Lavage with at least 750 mL of saline resulted in amelioration of local joint symptoms up to one year in an uncontrolled study [17].

Other studies however were unable to find a difference between needle aspiration and lavage [12, 13, 18]. Comparison of these contradictory studies is hampered by differences in study design and outcome measures.

We sought to evaluate retrospectively the potential therapeutic efficacy of arthroscopic lavage in a cohort of fifty patients with non-septic, inflammatory arthritis of a knee necessitating local treatment. The duration of effect after arthroscopic lavage was compared to the duration of effect after previous joint injection of the same knee in these fifty patients. We also evaluated potential predictive factors of clinical response.

(5)

Materials and methods Study design

We performed a retrospective analysis comparing the therapeutic effect of needle aspiration with arthroscopic lavage in the same patients treated at a single center (LUMC).

Chronologically, patients with an inflammatory arthritis of the knee underwent needle aspiration and after recurrence of arthritis they were eligible for arthroscopic lavage. We compared the time until recurrence of arthritis after needle aspiration with the time until recurrence after arthroscopic lavage. Patients served as their own control in comparing the time until recurrence of arthritis after both interventions.

Patient selection

The charts of all patients who underwent arthroscopic lavage of a knee between December 1998 and October 2001 were reviewed (n=86). Patients who had not been treated with needle aspiration prior to the arthroscopic lavage were excluded (n=36). Arthroscopy in these patients was done for diagnostic and/or trial purposes. Consequently, fifty patients with non- septic, inflammatory arthritis of a knee treated with needle aspiration followed by an arthroscopic lavage of the same knee were included in the present analysis.

Patient's characteristics

Patient's characteristics are shown in Table 1. Fifty patients were evaluated of whom 29 were men (58%) and 21 were women (42%) with a mean age of 58.8 years (range 44-81, s.d. 9.1) at the time of arthroscopic lavage. Thirty-six patients had a diagnosis of rheumatoid arthritis (RA), ten patients had mono- or oligo-arthritis of undifferentiated origin (AUO) and four patients suffered from spondylarthropathy (SpA). Mean disease duration was 8.3 years (range 0.1-40 yrs., s.d. 9.0). Thirty-seven patients (74%) used DMARDs with a mean total number of (previous plus current) DMARDs of 2.5 (range 0-7, s.d. 2.1). The mean number of previous joint injections, including the last one prior to arthroscopy, was 3.1 (range 1-13, s.d. 2.9).

Thirty out of our fifty patients had erosive disease, (29 RA, 1 SpA) and all twenty-three rheumatoid factor positive patients had RA. Radiographic scores were based on non- standardised anteroposterior knee radiographs according to Kellgren's score (0-4) for osteoarthrotic signs [19]. The mean score in the whole cohort was 1.40 (range 0-4, s.d. 1.15).

At needle aspiration twelve patients did not have steroids injected while at arthroscopy four times no steroids were applied. The reasons for not applying steroids were a diagnostic

(6)

aspiration or lavage in eight resp. four cases and lack of effect of steroids in former joint aspirations in the remaining four cases.

Needle aspiration

After maximal aspiration of synovial fluid from the knee joint an injection with triamcinolone hexacetonide (AHP Pharma Wyeth Lederle BV, Hoofddorp, The Netherlands) and lidocaine 0.5% (3-5mL) was given. The mean administered dose of steroids was 23 mg (range 0-60, s.d.

16 mg). Patients were instructed to minimize physical activities involving the knee joint for at least two days.

Arthroscopic lavage

Arthroscopy was performed by experienced arthroscopists (MvO and JvL) using a small-bore 2.7 mm arthroscope (Storz, Tuttlingen, Germany) with sterile technique. After maximal needle aspiration 30 mL of 0.5 % lidocaine was applied through the same needle for intra- articular and skin anaesthesia. Inferolateral of the patella topical anaesthesia was achieved by 30 mL of 1% lidocain. Two small skin incisions were made to introduce two portals into the joint. The lower portal was used for introduction of the arthroscope and instillation of saline.

The upper portal was used for taking of synovial biopsies with a 2.0 mm forceps and draining of the saline. Lavage was performed by flushing the joint with at least 1000 mL of saline. At the end of the procedure 6-8 mL of 0.5% marcaine and a mean dose of 34 mg (range 0-60, s.d. 13 mg) of triamcinolone hexacetonide were administered. After removal of the two portals both skin incisions were approximated with steristrips. Patients were instructed to minimize physical activities involving the knee joint for at least two days.

Outcome measures

The primary and single outcome measure after both interventions was defined as symptomatic recurrence of knee synovitis (swelling and/or effusion) assessed by the treating rheumatologist. All patients had recurrence of arthritis after needle aspiration since this was obligatory before arthroscopic lavage could take place.

Statistical analysis

The cumulative incidence of recurrences of knee synovitis for both treatments was determined by constructing Kaplan-Meier survival curves. A possible difference in recurrence rate

(7)

between the treatments was examined through a variant of the Cox proportional hazards regression model that allows for the participation of patients in both treatments by robust variance estimates [20]. Dichotomous variables were created for clinical characteristics (see Table 2). The Cox model was used to evaluate the effect of clinical characteristics on the recurrence rate. Furthermore, we assessed whether the recurrence rate after both treatments differed between categories of clinical parameters. The Cox model produces a regression coefficient for each covariate which can be interpreted as a relative risk (RR) when taken exponentially. Other statistical analyses were performed using Student's t-test and F² test where appropriate. All statistical analyses were done with the statistical software package STATA 6.0 (StataCorp, College Station, TX). Values of p<0.05 were considered statistically significant.

Results

The cumulative incidence of recurrence of arthritis at one year follow-up was 0.94 and 0.58 respectively for needle aspiration and arthroscopic lavage. The Cox regression analysis showed a 3.0-fold decrease (95% C.I. 2.1-4.4, p<0.001) of recurrence rate after lavage compared to needle aspiration. A Kaplan-Meier survival curve for the duration of effect after both interventions is plotted in Figure 1. The mean and median duration of effect after joint aspiration was 4.7 (range 0.03-77.6, s.d.11.6) and 1.9 months respectively. At the time of analysis twenty-three patients did not have recurrence of arthritis after lavage with a mean follow up time of 11.3 months. In twenty-seven patients who had recurrence of arthritis after lavage the duration of effect was 3.0 months (SD 2.3). The median time until recurrence after arthroscopic lavage was 4.8 months (no mean available because arthritis had not recurred in 23 patients). The time between needle aspiration and arthroscopy was 6.7 ± 12.1 months (mean ± SD) without differences between the diagnosis groups (data not shown). Patients with RA had significant longer disease duration, higher numbers of previously used DMARDs and more patients were on oral prednisone therapy compared to patients with other causes of knee inflammation (see Table 1). The administered dose of steroids was 23 mg at joint aspiration and 34 mg at lavage (p<0.001, Students t-test), with no differences beween diagnosis groups (pt0.27 for both interventions). In 16 patients DMARDs were changed between needle aspiration and lavage. Concomitant DMARD use at both interventions was comparable (p=0.50 with Chi-square test). Respectively sixteen and ten patients had a change

(8)

of DMARD use in the follow-up period after lavage and needle aspiration (p=0.02 with Chi- square test).

Clinical prediction of response

In search of clinical parameters predicting the outcome of arthroscopic lavage and needle aspiration we determined the relative risk for recurrence of arthritis of a reference category versus an index category as shown in Table 2. This table shows the effect of a clinical parameter on recurrence of arthritis after lavage and needle aspiration in all patients and in the RA-group. Patients with more previous DMARDs and longer disease duration were found to benefit more from lavage than patients from the other dichotomous categories. Kaplan-Meier survival curves for both interventions stratified for previous DMARD use are plotted in Figure 2.

Relative risk between interventions

The relative risks for recurrence of arthritis after arthroscopic lavage compared to needle aspiration within the strata of clinical parameters are shown in Table 3. There was at least a 1.5-fold decrease in recurrence rate after lavage as compared to needle aspiration which was statistically significant in all the strata except for short disease duration (d 2 yr) and application of no or a low dose of steroids at the intervention (d 20 mg). There was a significant difference in treatment effect between the reference and index category of previous DMARD-usage in all patients (4.5 vs. 1.8 for >1 vs. d1 with p=0.02) and for disease duration in all and RA-patients (4.5 vs. 1.5 and 5.7 vs. 1.6 resp. for > 2 and d 2 yr. with p=0.04 resp.

0.02).

Effect of steroids

We compared the duration of effect after joint aspiration and lavage for various amounts of steroids injected at the intervention. The duration of effect after joint aspiration (mean r s.d.) was 1.1 r 2.2, 7.8 r 17.4 and 3.8 r 5.1 months for respectively 0, 20 and 40 mg of triamcinolone hexacetonide injected. There were no statistically significant differences between groups. Cox regression analysis of the effect of steroids on each intervention seperately showed a better response to treatment when steroids were injected (HR 2.0 and 3.1 for lavage resp. aspiration; 0 vs. t 20 mg, only significant for aspiration with p=0.04).

(9)

However, there was no dosage effect of the amount of steroids injected (HR 1.3 and 0.8 for lavage resp. aspiration; 20 vs. 40 mg, both NS).

Side effects

Both procedures were well tolerated by patients. Some tenderness of the knee was reported by a minority of patients (7 out of 50) after arthroscopy. No bleeding or infectious complications occurred.

Discussion

In this analysis of patients with refractory inflammatory arthritis of the knee we found a statistically significant and clinically relevant difference in effect of arthroscopic lavage compared to needle aspiration. Patients had a three times greater risk of recurrence of arthritis within one year after aspiration compared to lavage. Cox proportional hazard regression analysis showed patients with more previous DMARDs and longer disease duration to benefit significantly more from lavage than patients from the other dichotomous category. Our results were not affected when adjusted for the injected amount of steroids although the application of steroids itself led to a longer duration of effect after both interventions compared to no steroids injected.

Some studies [16, 17] have shown a beneficial effect of arthroscopic lavage compared to needle aspiration although others failed to demonstrate superiority of arthroscopic lavage [12, 13, 18]. Comparison of these contradictory results is difficult because of different study designs and outcome measures. We think the larger amount of saline used for lavage in our study, which leads to more extensive removal of joint debris, is the main reason why we found a difference compared to several others. Few studies have explored clinical parameters predicting a favourable outcome of needle aspiration or arthroscopic lavage [21, 22]. Only a higher ESR in JIA-patients predicted a better outcome for joint injection of the knees [23]. In our study longer disease duration and more previous DMARDs were predictive clinical parameters for a positive outcome of arthroscopic lavage. Since the number of used DMARDs is strongly correlated with disease duration we assume disease duration is the determining predictive clinical parameter. In accordance with the lack of evidence in the literature, we did not find a dose-response effect of steroids after either intervention. However, this would contradict the hypothesis that the diminished dilution of injected steroids is responsible for the

(10)

beneficial effect of synovial fluid aspiration [5, 10]. Maybe the removal of joint debris is of greater importance than a higher intra-articular concentration of steroids.

Because of its retrospective design our study has some limitations of which the assessment of knee inflammation and the difference in the amount of applied steroids are the most obvious.

Some authors [24-27] have suggested semi-quantitative measurement for knee inflammation but in our study recurrence of arthritis was assessed on the basis of the need for local treatment as it occurs in daily practice. This method has been employed before because of its clinical relevance [4, 28, 29]. We assume a putative inequilibrium is balanced between both interventions. Although greater amounts of steroids were injected and a higher number of patients received steroids with lavage compared to aspiration, the lack of a dose response effect indicates steroids only had a limited contribution to the superior effect of lavage in our study. A frequent problem with joint injections is misplacement of the injection [30]. One study found as much as 9% of needle-injections for knee-arthritis to be extra-articular [31].

With arthroscopic techniques the steroids are placed within the joint under vision by experienced arthroscopists. Some of the relapses after aspiration may have been caused by inaccurate placement, but we don't know how many injections were extra-articular. Another difficulty is the lack of a control group. Patients and doctors are biased to expect more effect after a more aggressive intervention.

Our results show that patients with a longer disease duration responded better to arthroscopic lavage than the other patients. It is assumed that a patient with longstanding arthritis has suffered from inflammatory processes within the knee joint for years and the joint is filled with debris from former inflammation (fibrin, rice bodies). These degradation products sustain the inflammation [8,32-34] and can lead to further damage of the cartilage. Removal of fibrin by lavage could therefore diminish the degree of inflammation. The difference in effect after lavage between RA and non-RA patients (RR 1.5, p=0.33) may be explained by the shorter disease duration in non-RA patients. In non-RA patients however there is still a 1.9 times greater chance of recurrence after joint aspiration compared to lavage. This suggests not only the removal of fibrin (after years of inflammation) but the more extensive removal of inflammatory cells and degradative enzymes might contribute to a decrease in inflammation as well.

In this study we showed a superior effect of arthroscopic lavage compared to needle aspiration with a three times longer duration of effect. We compared both interventions in fifty patients with several causes of knee arthritis and found an intrinsic therapeutic effect of

(11)

lavage independent from the application of steroids. Since this study was biased by its retrospective design we need to confirm these findings in a prospective controlled study.

Arthroscopic lavage is a safe and well tolerated outpatient procedure with therapeutic advantages especially for patients with inflammatory arthritis of the knee refractory to joint injections.

Acknowledgements

We thank Drs. R.J. Verburg for technical assistance in the arthroscopic procedures.

(12)

References

1. Hollander JL, Brown EM, Jessar RA, Brown CY. Hydrocortisone injected into arthritic joints: comparative effect of and use of hydrocortisone as local anti-arthritic agent. JAMA 1951;147:1629-35

2. McCarty DJ. Treatment of rheumatoid joint inflammation with triamcinolon hexacetonide. Arthritis Rheum. 1972;15:157-73

3. Chakravarty K, Pharoah PDP, Scott DGI. A randomised controlled study of postinjection rest following intra-articular steroid therapy for knee synovitis. Br J Rheumatol 1994;33:464-8

4. Blyth A, Hunter JA, Stirling A. Pain relief in the rheumatic knee after steroid injection: a single blind comparison of hydrocortisone succinate, and triamcinolone acetonide or hexacetonide. Br J Rheumatol 1994;33:461-463

5. Weitoft T, Uddenfeldt P. Importance of synovial fluid aspiration when injecting intra- articular corticosteroids. Ann Rheum Dis 2000;59:233-235

6. Burman MS, Finkelstein H, Mayer L. Arthroscopy of the knee joint. J Bone Joint Surg 1934;16:255-68

7. Jayson MIV, Dixon AStJ. Arthroscopy of the knee in rheumatic diseases. Ann Rheum Dis 1968;27:503-11

8. Popert AJ et al. Frequency of occurence, mode of development and significance of rice bodies in rheumatoid joints. Ann Rheum Dis 1982;41:109-17

9. Baeten D. et al. Needle arthroscopy of the knee with synovial biopsy sampling:

Technical Experience in 150 patients. Clin Rheum 1999;18:434-41

10. Neustadt DH. Synovitis of the knee. Effects of postinjection rest. Clin Rheum Pract 1985;3:65-8

11. Fassbender HG. Rheumatoid arthritis. In: Pathology of rheumatic diseases.

Berlin:Springer-Verlag. 1975;79-210

12. Bird HA, Ring EFJ. Therapeutic value of arthroscopy. Ann Rheum Dis 1978;37:78-9 13. Lindsay DJ, Ring EFJ, Coorey PFJ, Jayson MIV. Synovial irrigation in rheumatoid

arthritis. Acta Rheum Scand 1971;17:169-174

14. Jayson MIV, Dixon AStJ. Intra-articular pressure in rheumatoid arthritis of the knee.

Ann Rheum Dis 1970;29:261-5

15. Dixon ASJ, Emery P. Local injection therapy in rheumatic diseases. 4th ed. Basel:

Eular Publishers 1992

(13)

16. Fitzgerald O, Hanly J, Callan A, McDonald K, Molany J, Bresnihan B. Effects of joint lavage on knee synovitis in rheumatoid arthritis. Br J Rheum 1985;24:6-10

17. Sharma A et al. Arthroscopic lavage treatment in rheumatoid arthritis of the knee. J Rheumatol 1996;23:1872-4

18. Srinivasan A, Amos M, Webley M. The effect of joint washout and steroid injection compared with either joint washout or steroid injection alone in rheumatoid knee effusion. Br J Rheum 1995;34:771-3

19. Altman RD, Hochberg M, Murphy WA, Wolfe F, Lequesne M. Atlas of individual radiographic features in osteoarthritis. Osteoarthritis Cartilage 1995;3:3-70 20. Lin, D. Y., and L. J. Wei. The robust inference for the Cox proportional hazards

model. J. Am. Stat. Assoc. 1989;84: 1074-8 .

21. Jones A, Doherty M. Intra-articular corticosteroids are effective in osteoarthritis but there are no clinical predictors of response. Ann Rheum Dis 1996;55:829-32 22. Luukkiainen R, Hakala M, Sajanti E, Huhtala H, YliKerttula U, Hämenkorpi R.

Predictive value of synovial fluid analysis in estimating the efficacy of intra-articular corticosteroid injections in patients with rheumatoid arthritis. Ann Rheum Dis 1992;51:874-6

23. Ravelli A, Manzoni SM, Viola S, Pistoria A, Ruperto N, Martini A. Factors affecting the efficacy of intraarticular corticosteroid injection of knees in juvenile idiopathic arthritis.J Rheumatol 2001;28:2100-2

24. Nicholas JJ, Taylor FH, Buckingham RB, Ottonello D. Measurement of circumference of the knee with ordinary tape measure. Ann Rheum Dis 1976;35:282-3

25. Paterson J. et al. Assessment of rheumatoid inflammation in the knee joint. Ann Rheum Dis 1978;37:48-52

26. Marks JS, Palmer MK, Burke MJ, Smith P. Observer variation in the examination of knee joints. Ann Rheum Dis 1978;37:376-7

27. Kirwan JR, Byron MA, Winfield J, Altman DG, Gumpel JM. Circumferential measurement in the assessment of synovitis of the knee. Rheum Rehab 1979;18:78-84 28. Jalava S, Virsiheimo B. Triamcinolone hexacetonide in the treatment of therapy-

refractory gonitis. J Int Med Res 1982;10:53-8

29. Padeh S, Passwell H. Intraarticular corticosteroid injection in the management of children with chronic arthritis. Arthritis Rheum 1998;7:1210-4

(14)

30. Jones A, Regan M, Ledingham J, Pattrick M, Manhire A, Doherty M. Importance of placement of intra-articular steroid injections. BMJ 1993;307:1329-30

31. Bliddal H. Placement of intra-articular injections verified by mini air-arthrography.

Ann Rheum Dis 1999;58:641-6

32. Dumonde DC, Glynn LE. The production of arthritis in rabbits by an immunological reaction to fibrin. Br J Exp Pathol 1962;43:373-83

33. Glynn LE. The chronicity of inflammation and its significance in rheumatoid arthritis.

Ann Rheum Dis 1968;27:105-21

34. Scott DL, Wainwright AC, Walton KW, Williamson N. Significance of fibronectin in rheumatoid arthritis and osteoarthritis. Ann Rheum Dis 1981;40:142-53

(15)

Table 1

Patient's characteristics

Diagnosis

Number of patients (males)

RA

n=36 (23)

AUO

n=10 (4)

SpA

n=4 (1) Mean age in yr (s.d.) 60.2 (9.3) 56.3 (7.2) 52.4 (9.7)

range 44-81 47-70 44-64

Mean disease duration in yr (s.d.) 10.4 (9.6) 1.3 **(1.2) 6.8 (4.8)

range 0-40 0-3 3-14

Mean no. of previous DMARDS (s.d.) 3.2 (2.0) 0.3 **(0.7) 1.3 (1.3)

range 0-7 0-2 0-3

Mean no. of previous joint inj. (s.d.) 3.3 (3.1) 1.9 * (1.0) 4.0 (3.4)

range 1-13 1-3 2-9

Concomitant DMARD use (%) 33 (92) 2 (20) 3 (75)

Patients on oral prednisone 12 1 0

Mean prednisone dose (mg/d) 11.1 7.5* 0

Radiologic knee score 1.4 1.0 2.3

RA=rheumatoid arthritis, AUO=Arthritis of Unknown Origin, SpA=spondylarthropathy.

*=p<0.05 (Student's t-test AUO vs. RA) **=p<0.001 (Student's t-test AUO vs. RA)

(16)

Table 2. Relative risk for recurrence of arthritis between two categories of a clinical parameter after arthroscopic lavage and needle aspiration in all and RA-patients.

All (n=50) RA (n=36)

Parameter Reference/Index

Relative Risk

p-value Relative Risk

p-value

lavage 2.6 0.01 2.4 0.08

Previous DMARDs

nd1 / n>1 aspiration 1.1 0.79 1.1 0.87

lavage 2.2 0.046 2.1 0.10

Disease duration

d 2 yr / > 2 yr aspiration 0.8 0.53 0.6 0.17

lavage 1.3 0.47 0.8 0.54

Previous injections

nd2 / n>2 aspiration 0.9 0.72 0.7 0.19

lavage 2.1 0.09 1.2 0.78

Concomitant DMARD

no / yes aspiration 1.2 0.55 1.8 0.12

lavage 1.6 0.28 1.1 0.78

Oral prednisone no / yes

aspiration 0.8 0.38 0.8 0.40

lavage 1.3 0.45 1.2 0.71

Radiologic score¹⁾ 0-1 / 2-4

aspiration 1.1 0.71 0.8 0.86

lavage 1.5 0.33

Diagnosis nonRA/RA

aspiration 0.9 0.81

lavage 2.0 0.53 ³⁾ ³⁾

Steroid²⁾ dose

0*mg / t20mg aspiration 3.1 0.04 1.8 0.40

lavage 1.3 0.57 1.9 0.17

Steroid²⁾ dose 20mg / 40mg

aspiration 0.8 0.51 0.7 0.23

1)Kellgren's score for osteo-arthrotic signs ²⁾Triamcinolone hexacetonide ³⁾Too small sample sizes for meaningful analysis *4 and 12 patients had 0 mg of steroids injected at lavage resp.

needle aspiration in all patients and 2 versus 6 resp. in RA patients

(17)

Table 3. Relative risks for recurrence of arthritis after needle aspiration compared to arthroscopic lavage in all and RA patients for various clinical parameters.

All (n=50) RA (n=36)

Relative

Risk p- value

number of patients

Relative Risk

p- value

number of patients

Total 3.1 <0.0 50 3.7 <0.00 36

nd1 1.8 0.02 19 2.2 0.21 7

d2 yr. 1.5 0.20 19 1.6 0.33 11

Disease

duration > 2 yr. 4.5 <0.0 31 5.7 <0.00 25

nd2 2.5 0.001 22 3.6 0.002 20

no 1.9 0.04 12 8.2 0.01 3

Concomitan

t yes 3.6 <0.0 38 3.4 <0.00 33

no 2.5 <0.0 37 3.2 <0.00 ²⁴

Oral

prednisone yes 8.8 <0.0 13 7.8 <0.00 12

0-1 2.8 <0.0 25 3.1 0.001 18

Radiologic

Score¹⁾ 2-4 3.5 <0.0 25 4.8 <0.00 18

RA 3.7 <0.0 36

Diagnosis

nonRA 1.9 0.02 14

0*mg 2.4 0.34 12 vs 4^{# 3)} ³⁾ 6 vs 2^# 20mg 1.9 0.15 20 vs 7# 1.6 0.27 17 vs 6^# Steroid²⁾

dose

40mg 3.8 <0.0 18 vs 39# 5.9 <0.00 13 vs 28^#

1)Kellgren's score for osteo-arthrotic signs ²⁾Triamcinolone hexacetonide³⁾Too small sample sizes for meaningful analysis ^#number of patients, aspiration vs lavage

(18)

Figure 1. Kaplan-Meier survival curves for the recurrence of arthritis after lavage (bold) and aspiration in all patients (n=50).

(19)

Figure 2. Kaplan-Meier survival curves for the recurrence of arthritis after lavage (dotted) and aspiration for patients with >1 previous DMARD (bold) or d1 previous DMARD.