British Journal of Rheumatology 1994;33:461-463

PAIN RELIEF IN THE RHEUMATOID KNEE AFTER STEROID INJECTION

A SINGLE-BLIND COMPARISON OF HYDROCORTISONE SUCCINATE,
AND TRIAMCBNOLONE ACETONIDE OR HEXACETONIDE
T. BLYTH, J. A. HUNTER and A. STIRLING
Department of Rheumatology, Gartnavel General Hospital, Glasgow

SUMMARY
Since the introduction of intra-articular steroid therapy 40 yr ago there have been many changes in the treatment of rheu-
matoid patients. Previous studies suggest differing times of response for the same agents. This study reports the response,
measured by a five-point pain chart, of 300 patients with painful rheumatoid knees. Sixty received hydrocortisone succi-
nate (HC), 150 received triamcinolone acetonide (TA), and 120 triamcinolone hexacetonide (TH).
Results demonstrated little effect with HC, but good responses with 2TA and TH. More patients were rendered painfree
for a longer time with TH; 18% at 12 weeks, as against 9% with TA (x test P<0.005). At 12 weeks 59% showed continued

Downloaded from http://rheumatology.oxfordjournals.org/ at University of Winnipeg on September 16, 2014

improvement with TH as against 44% with TA (x2 test P<0.05).
TH is the preferred preparation for injection of the rheumatoid knee.
KEY WORDS: Rheumatoid arthritis, Knee joint, Pain relief, Intra-articular injection, Hydrocortisone succinate, Triam-
cinolone acetonide, Triamcinolone hexacetonide.

THE British National Formulary (BNF) [1] lists several
corticosteroid preparations for the intra-articular
treatment of the rheumatoid knee. Previous studies
have suggested widely differing durations of response
following injection of the same agent [2]. For instance
the action of hydrocortisone acetate is stated as lasting
a mean of 6 days in one report [2] and 40 days in
another [3]. Similarly triamcinolone hexacetonide
(TH) lasts 21 [2], 59 [3] or 90 [4] days in different stud-
ies. Triamcinolone acetonide (TA) has been reported
to have a mean duration of response of 14 [2] days, but
our experience suggests more prolonged benefit. One
explanation may be that other systemic treatments
have evolved since the original studies and better
underlying control of the disease enables a more pro-
tracted response. This paper seeks to clarify and com-
pare the duration of pain relief in a large cohort of
rheumatoid patients receiving intra-articular treat-
ment for a knee joint exacerbation. The agents studied
were chosen by virtue of the availability of hydro-
cortisone succinate (HC) in most clinical settings, the
premier position of TH in the BNF [1], and the wide-
spread use of TA.
PATIENTS AND METHODS
Three hundred patients with RA [5] and trouble-
some knee symptoms were recruited. They either had
pain and stiffness accompanied by features of active
synovitis on examination, or a history of previous
response to intra-articular steroid, or both. Patients
were excluded if they had commenced or changed their
second-line drug in the preceding 3 months. No
patients had received systemic steroid or had received
an injection in the treated joint in the preceding 3
months. Intra-articular injection of another joint was
not permitted within 4 weeks of the study injection.
Immediately before injection, effusion, thickening, and
tenderness were recorded on four-point scales. ESR,
haemoglobin, platelet count and RF titres were
checked. Patients were requested to record their pain
using a five-point pain scale weekly for 12 weeks. With-
drawal from the study was permitted if symptom con-
trol proved inadequate and further intra-articular
steroid was administered to the same joint.
Injections were performed at an outpatient clinic
using aseptic precautions, a 'no touch' technique, and
local anaesthesia with 2% lignocaine to skin and sub-
cutaneous tissues. Aspiration was attempted using a 19
gauge needle and the volume of fluid obtained
recorded. Aspirate was analysed bacteriologically, for
white cell count, glucose and RF (where sufficient
aspirate was obtained).
The first 60 patients were randomized [6] to receive
either 100 mg intra-articular HC or 40 mg intra-artic-
ular TA. The remaining 240 patients similarly received
at random either 40 mg TA or 20 mg TH. Statistical
analysis was non-parametric using Mann-Whitney U
test, x2 tests and Spearman Rank Correlation.
RESULTS
The table shows baseline data for the treatment
groups. There were no differences in measures of effu-
sion, thickening, tenderness, haemoglobin, platelet
count, or RF. The only difference demonstrated was
the higher volume of aspirate obtained in the hydro-
cortisone group. The reason for this is not clear and it is
suspected that the difference occurred by chance.

Department of Rheumatology, Gartnavel General Hospital, Drop-outs

Glasgow. More patients who received HC required further
Submitted 9 March 1993; revised version accepted 4 January treatment to the injected knee than those who received
1994. either preparation of triamcinolone. Only 11 of 30

0263-7103/94/050461 + 03 $08.00/0 © 1994 British Society for Rheumatology

461
462 BRITISH JOURNAL OF RHEUMATOLOGY VOL. 33 NO. 5

response, assessed as area under the curve, there were

significant correlations with age in the TA group
(Spearman Rank correlation coefficient = 0.2, P<0.05)
and in baseline platelet count in the TH group (Spear-
man Rank correlation coefficient = 0.169, P<0.05).
I The correlation coefficients suggest these are unlikely
I to have predictive value.
Adverse events
f» No patient reported an adverse reaction to the injec-
tion. One week after injection, one patient in the TA
group and two patients in the TH group reported
deterioration. At 4, 8 and 12 weeks, 4, 6 and 8%
reported deterioration in the TA group. In the TH
group thefigureswere 3,4 and 7%. No patient suffered

Downloaded from http://rheumatology.oxfordjournals.org/ at University of Winnipeg on September 16, 2014

5 6 7 chemical synovitis sufficient to warrant additional
Weeks after injection
arthrocentesis.
FIG. 1.—Proportion of patients remaining pain-free after injec- DISCUSSION
tion. g|, Triamcinolone acetonide; • , triamcinolone
hexacetonide.
These studies show that the benefits of intra-artic-
ular TH exceed those of TA, and that the readily avail-
patients who received HC achieved follow-up beyond 4 able preparation of HC is substantially less effective.
weeks, and at this stage only four rated their pain less Despite the large numbers studied there was no clear
severe than before injection. None had gained com- pretreatment measurement that predicted a more
plete pain relief at anytime. For these reasons no favourable response. This study adds to the previous
further graphical representation of efficacy data for reports of benefit from intra-articular steroid and
HC is included. characterizes the types of patients being treated. The
Twenty-two (14.5%) patients in the TA group and 13 aim was to recruit a representative sample from a rou-
(11%) patients in the TH group failed to return pain tine injection clinic. The indications for injection were
scores. Two patients both from the TH group were mainly defined by the patients' symptoms or requests;
excluded, one because of starting second-line therapy this reflects the aim of intra-articular treatment to
shortly after the injection and another because of hos- relieve pain.
pital admission for trauma to the contralateral ankle. Previous pharmacokinetic studies [7] have shown
There were no significant differences in the drop-out that the total amount of triamcinolone absorbed into
rate between the TA and TH groups. the circulation after intra-articular injection did not dif-
fer between TA and TH. However the rate of absorp-
Efficacy tion was more rapid with TA and the slower release of
At week 4 there were significantly better pain scores TH was thought to be related to its lower solubility. No
in the TH group compared with the TA group delay in benefit of TH was demonstrated in this study.
(X2 = 14.5, P = 0.0023). The superior pain relief of TH
was sustained for the remainder of the follow-up TABLE I
period. Clinical entry characteristics of patients
The figure shows the proportion of patients who Hydrocortisone Triamcinolone Triamcinolone
gained complete relief of pain at the treated knee. succinate acetonide hexacetonide
Again TH proved significantly superior to TA by week
4 (x2 = 12.6, P<0.0005) and the difference was sus- No. 30 150 120
tained until week 12. Age range (yr) 24-76 24-79 31-78
Percent female 72 83 80
When response to therapy was defined as a fall of Pain
one grade or more on the pain scale, 87% of TA and None 0 0 0
90% of TH treated patients improved. This response Mild 5 (17%) 16 (11%) 14 (12%)
was sustained to 4 weeks in 76%, 8 weeks 63%, and 12 Moderate 7 (23%) 37 (25%) 33 (27%)
Severe 13 (43%) 74 (49%) 52 (44%)
weeks in 44% of TA-treated patients and the compar- Very severe 4 (13%) 23 (15%) 20 (17%)
able figures for the TH-treated group were 80% at 4
weeks, 72% at 8 weeks, and 59% at 12 weeks. The Median 60 50 60
difference between the treatment groups reached sig- Range 6-108 3-140 1-140
nificance at only week 10 (x2 test P<0.05) however. Fluid obtained from the knee
Percent 83% 71% 67%
Analysis of data using the method of 'area under the patients
curve' confirmed the greater duration of pain relief in Amount (ml) 20.5* 10» 8*
the TH group (median for TA was 13 CI 11.5-14.5; and (mean)
for TH was 9.5 CI 7.5-11 P = 0.0029 Mann-Whitney U « P<0.05 >
, ~P<0.005 »
test).
When baseline measures were correlated with •Statistical difference Mann-Whitney U test.
 BLYTH ETAL.: RHEUMATOID KNEE AND STEROID INJECTION
The brief duration of benefit of HC may relate to this
being a soluble preparation.
Whilst the placebo response to injection techniques
might be substantial, the differences between hydro-
cortisone and triamcinolone preparations in this report
suggest a genuine pharmacological effect. A mech-
anism that explains the prolonged pain relief after TH
injection remains to be defined as measurable TH is
only detectable in the venous blood up to approxi-
mately 4 weeks after injection.
This study was not designed to investigate adverse
drug reactions, but volunteered symptoms were rare
(1% immediate knee symptom exacerbations in tri-
amcinolone-treated patients). No cases of infection
occurred. Crystal synovitis due to intra-articular depot
steroid preparations is described [8, 9]. In this large
study, whilst five patients who received TH appeared
to improve gradually to total loss of pain over 4 weeks,
no instances of acute synovitis after injection were
encountered.
The duration of response to intra-articular steroids
has proved variable in other studies. Most reports date
from the 1950s and 1960s when the use of second-line
drugs and the variety of NSAIDs would have been less
than at present [2, 3]. In addition the use of systemic
corticosteroids or ACTH might be expected to reduce
the impact of a local steroid injection. Alternatively
previous studies may have used different measures to
assess response. Data in this study is consistent with
one report of prolonged clinical, radioisotope and ther-
mographic response to intra-articular steroids [10].
Previous studies have also suggested more prolonged
relief with TH compared with methyl prednisolone [11]
and hydrocortisone acetate [3].
This study confirms the previously described symp-
tom relief afforded by intra-articular triamcinolone
and gives useful data for patient information on antici-
pated duration of pain relief. We hope this will prove
valuable to those informing their patients about joint
463
injection. Hexacetonide is the preferred preparation of
triamcinolone for injection of painful rheumatoid
knee.
REFERENCES
1. Prasad AB (ed.) Local corticosteroid injections. In:
British National Formulary. London: British Medical
Association and the Pharmaceutical Press; 1993;26:
359-60.
2. Hollander JL. Intra-synovial corticosteroid therapy in
rheumatoid arthritis. MD Med J 1970;19:62-6.
3. Rigby P, Glick EN, Smith R. Intra-articular injection
of triamcinolone in rheumatoid arthritis. Proceedings
of the VHth European Rheumatology Congress,
Brighton 1971:4-13.
Downloaded from http://rheumatology.oxfordjournals.org/ at University of Winnipeg on September 16, 2014
4. Anttinen J, Oka M. Intra-articular triamcinolone hex-
acetonide and ostnic acid in persistent synovitis of the
knee. Scand J Rheumatol 1975;*125-8.
5. Arnett FC, Edworthy SM, Bloch DA et al. The 1987
revised ARA criteria for classification of rheumatoid
arthritis. Arthritis Rheum 1988^1:315-24.
6. Random numbers. In: Documenta Geigy, 7th edn.
Macclesfield: Geigy Pharmaceuticals;1970:131.
7. Derendorf H, Mollmann H, Gruner A, Haack D,
Gyselby G. Pharmacokinetics and pharmacodynamics
of glucocorticoid suspensions after intra-articular
administration. Clin Pharm Ther 1986;39-313-17.
8. McCarty DJ, Hogan JM. Inflammatory reaction after
intra-articular injection of microcrystalline adreno-
corticosteroid esters. Arthritis Rheum 1964;7:359.
9. Berger RG, Yount WJ. Immediate 'steroid flare' from
intraarticular triamcinolone hexacetonide injection.
Arthritis Rheum 1990;33:1284-6.
10. De Silva M, Salisbury R, Hazleman BL, Parr G,
Raight P. Intra-articular steroid therapy in the rheu-
matic diseases: a clinical, radioisotopic and thermo-
graphic comparison of five steroid esters. Rev Rheum
Mai Osteoartic 1981 (June suppl) Abstracts: XVth
International Congress of Rheumatology.
11. Kendall PH. Triamcinolone hexacetonide a new
corticosteroid for intra-articular therapy. Ann Phys
Med 1967;<fc55.