KNEE EFFUSION AND REFLEX INHIBITION OF THE

QUADRICEPS
A BAR TO EFFECTIVE RETRAINING

H. FAHRER, H. U. RENTSCH, N. J. GERBER, Ch. BEYELER, Ch. W. HESS, B. GRUNIG

From the University of Bern, Switzerland

In order to investigate the difficulty of quadriceps training in the presence of an effusion into the knee we
examined 13 patients with chronic effusions by recording isometric muscle strength. Maximal strength was
markedly lower in the presence of an effusion, and aspiration of the effusion produced a 13.6% increase in
strength (p < 0.01). A further, small increase of 8% was recorded after intra-articular lignocaine injection.
Isometric strength and surface integrated EMG correlated well in six patients.

Two reflex mechanisms seem to inhibit quadriceps innervation in the presence of a persistent knee
effusion, one mediated by pressure sensitive receptors, the other still unknown. Joint aspiration and systemic
or intra-articular anti-inflammatory drug treatments are advised before any programme of quadriceps
training to allow maximum effects to be achieved.

Wasting of the quadriceps due to chronic knee disease is joint effusion. Intra-articular lignocaine abolished this
a common problem in rheumatology, orthopaedic sur- decrease.
gery and sports medicine. Both immobilisation and Both intra-articular fluid volume and joint angle
inflammation of the joint are considered to contribute influence intra-articular pressure. The capacity of a joint,
(Harding 1929; Staudte 1981). as with a rubber balloon, can be defined by a
Earlier workers have demonstrated a near-linear pressure/volume relationship. This is least in a position
relationship between volume and pressure of intra- of partial flexion, that is, from 30 to 60#{176}
(Eyring and
articular fluid under experimental and pathophysiolo- Murray 1964; Stratford 1981 ; Levick 1983 ; Machan
gical conditions (Jayson and Dixon 1970; Gerber and 1983). Quadriceps exercises, however, are usually done
Dixon 1974; McCarty 1980; Levick 1983). The reflex in full extension, when the pressure/volume relationship
inhibition ofthe quadriceps due tojoint effusion was first is raised (Basmajian 1970; Andrews 1982; Levick 1983).
demonstrated experimentally by de Andrade, Grant and It has been shown in experimental and postopera-
Dixon (1965). In 1984 Spencer, Hayes and Alexander tive effusions that the removal of fluid is followed by a
demonstrated a linear decrease of Hoffman’s reflex in fall in intra-articular pressure and a rise in the strength of
proportion to intra-articular fluid volumes in healthy quadriceps contraction (Bittscheidt, Hofmann and
subjects. Hoffman’s reflex is an electrically evoked Schumpe 1978 ; Kennedy, Alexander and Hayes 1982).
monosynaptic stretch reflex of the quadriceps, produced One recent study concluded that this phenomenon occurs
by low voltage stimulation of the femoral nerve. The only in cases of acute, but not chronic effusion (Jones,
threshold for inhibition was 20 to 30 ml of experimental Jones and Newham 1987).
In our study we tested the hypotheses that effusions
caused by arthritis inhibit quadriceps strength, that their
H. Fahrer, MD
H. U. Rentsch, MD removal increases quadriceps strength, and that physio-
N. J. Gerber, MD therapeutic exercises without prior removal of fluid are
Ch. Beyeler, MD
Department of Rheumatology of doubtful value.
Ch. W. Hess, MD
Department of Neurology
B. Gr#{252}nig,MD
PATIENTS AND METHODS
Department of Orthopaedic Surgery, University of Bern, Inselspital,
CH-3010 Bern, Switzerland.
We studied 13 patients (12 ofthem men) with a mean age
Correspondence should be sent to Dr H. Fahrer.
of 31 .3 years (range 16 to 57 years). In nine cases one
© 1988 British Editorial Society of Bone and Joint Surgery
030l-620X/88/4103 $2.00 knee was affected and in four cases, both knees. The
J Bone Joint Surg [Br] 1988 ;70-B :635-8. diagnosis was rheumatoid arthritis in three, Reiter’s

VOL. 70-B, No. 4, AUGUST 1988 635

636 H. FAHRER, H. U. RENTSCH, N. J. GERBER, CH. BEYELER, CH. W. HESS, B. GRUNIG

syndrome in two, undefined monarthritis in four and one patient, it irritated the knee and inhibited full muscle
degenerative joint disease in four. The effusions were all contraction. To allow the patients to adapt to the
chronic. procedure, two recordings of eight seconds were made

Muscle strength measurements. These were recorded from the control leg during maximal voluntary contrac-
tion. Recordings were then made from the affected leg,
using a Cybex II dynamometer (Cybex, Division of
before and after fluid aspiration and, in six patients, after
Lumex, Inc., New York, USA). The patients were sitting
intra-articular lignocaine.
with their abdomens and thighs tightly attached to the
chair and the knee fixed to the dynamometer in 60#{176}
of Evaluation. Planimetry was performed on the isometric
flexion. This angle was chosen because it gives a lower strength curves, using a Digiplan (Kontron Inc.); the
intra-articular pressure than further flexion or extension integral under the curve was determined for contraction
(Levick 1983). The strength of isometric extension was times of one and three seconds. The amplitude was
first recorded on the healthy or less affected side. measured after one and three seconds and at the maximal
Patients were asked to push with maximal strength for peak. Statistical evaluation of the strength changes was
eight seconds against the padded and blocked lever of the made by a paired two tailed t-test, with p < 0.05 as the
dynamometer. Results were recorded in Newton seconds significance threshold. The linear correlation coefficient
(Nsec), that is the product of strength and recording (r) and Spearman’s rank significance correlation coeffi-
time. cient (RS) were calculated.

Electromyography. In six patients an electromyogram

(EMG) ofthe medial vastus muscle was recorded on both RESULTS
sides in addition to the muscle strength. Two surface
electrodes were taped 6 cm apart over the belly and near
the tendon of the muscle, and the EMG signal was Of the various parameters of the strength curve, the
amplified by a DISA 1500 System (bandpass 10 Hz to integral area over three seconds (S3s) proved to be the
10 kHz) and subsequently processed by a DISA Mean most informative and this alone is reported as the
Voltage Unit Type l4C 20 to give a continuous plot of the parameter for maximal strength of contraction (Fig. 1).
numerical mean voltage. Isometric strength on affected and unaffected or less
Joint aspiration. This was performed after the initial set affected side. The two consecutive recordings of maximal
of measurements. A sterile 18 G single way needle isometric strength in the healthy or “better” leg showed
(0.8 mm lumen) was inserted under the upper medial identical results, the mean S3s values being 478 ±
pole of the patella without local anaesthesia. The mean 148 Nsec (Mean ± SD) for the first set of measurements
volume aspirated was 52.5 ± 25.6 ml (range 25 to and 482 ± 1 74 Nsec for the second one, showing
1 10 ml). In six subjects 5 ml of 2% lignocaine was satisfactory reproducibility and excluding a significant
injected at the same site after the second and before a learning or training effect (Fig. 2, left). As expected, the
third set of measurements. We did not use an indwelling maximal strength of the affected leg was significantly
plastic vein catheter since, in a preceding pilot study of lower 301 ± 203 Nsec, p<O.0S.

my

2
1

Integral: 1.31 mVsec Integral : 2.05 mVsec

N N

200 200

100 100

1 2 3 4 5 sec I 2 3 4 5 sec

before Aspiration after Aspiration

Fig. 1

Recordings from a typical subject, before and after aspiration of 80 ml of fluid, showing mean voltage and
integral ofthe EMG ofvastus medialis (above) and isometric strength (below). Maximal strength is usually
reached within three seconds of maximum voluntary activity.

THE JOURNAL OF BONE AND JOINT SURGERY

 KNEE EFFUSION AND REFLEX INHIBITION OF THE QUADRICEPS 637

Joint aspiration and intra-articular anaesthesia. None of seconds (p < 0.002). When an alteration of 10% or more
the patients experienced pain on quadriceps contraction of both the integrated strength and the EMG activity
either before or after aspiration. There was a significant was seen after aspiration, it was always in the same sense
increase of strength from 301 ± 203 to 342 ± 192 after for the two measurements. That is, when there was a
aspiration (Fig. 2, right). This increase was in the order clear-cut increase in muscle strength after aspiration, an
of 1 3.6%, the variability coefficient being 15%. increase in EMG voltage was also observed.
Six subjects who were given intra-articular anaes-
thesia immediately after the post-aspiration measure-
DISCUSSION
ment, had a third recording done five minutes after
injection of lignocaine. In all but one subject there was a We measured quadriceps strength in patients with
further increase in strength. In Subject I, the exception, clinical effusion before and after aspiration ofjoint fluid.
there was a significantly delayed increase in strength, The results have a bearing on why quadriceps training is
with poor results after three seconds, but much better often ineffective where there is a persistent effusion. Our
results after five seconds, of the same order as the three- results were reproducible and demonstrated a mean
second results in the other five subjects (see Fig. 2). The increase in strength of 1 3.6% after aspiration of about
whole group evaluation, including Subject I, showed 50 ml of fluid, with a further mean increase of 8% after
equal quadriceps strength before and after lignocaine, intra-articular lignocaine injection. The integrated sur-
but, excluding Subject I, the other five showed a face EMG activity of the medial vastus, measured in
significant increase in strength of 8% (p.ezO.Ol). There some patients on both sides, correlated well with
was no clear correlation between strength changes and quadriceps strength, confirming previous findings that
volume of joint fluid. EMG gives a good estimate of muscle force, though it
Maximal strength and surface EMG. The comparison of provides relative rather than absolute values (de Vries
muscle strength and the mean voltage of the EMG 1968; Moritani and de Vries 1978; Stratford 1981;
recorded simultaneously in six patients showed a linear Soderberg and Cook 1983). The fact that EMG activity
correlation between the two parameters when the also increased when there was a clear-cut gain in muscle
integrated value of each was taken over one and three strength after aspiration supports neurogenic rather than

N sec
n.s. p<0,01 n.s.
800
(p0.01 without
subject #{174})

600

400

200

---- mean value

Fig. 2

Maximal isometric strength of quadriceps. On the left, values for the first and second attempts with the
normal
six subjects,
or less affected
after
le are given.
the injection
On the
of lignocaine
right are values
A.
before aspiration #{149}
after aspiration 0 and, for

VOL. 70-B, No. 4, AUGUST 1988

638 H. FAHRER, H. U. RENTSCH, N. J. GERBER, CH. BEYELER, CH. W. HESS, B. GRUNIG

mechanical joint inhibition. Thus, our observations de Andrade JR, Grant C, Dixon AStJ. Joint distension and reflex
muscle inhibition in the knee. J Bone Joint Surg [Am]
confirm and quantify the postulate ofan inhibitory reflex
1965 ;47-A :313-22.
mechanism blocking voluntary quadriceps innervation Dee R. The innervation ofjoints. I. Structure ofjoint innervation. In:
(de Andrade et al. 1965). Sokoloff L, ed. The joints and synovia/ fluid. New York, etc:
Conclusions. Our experiments seem to demonstrate that Academic Press, 1978:177-83.

at least two different sorts of afferent impulses are the deVries HA. “Efficiency of electrical activity” as a physiological
measure of the functional state of muscle tissue. Am J Phys Med
cause of this muscle inhibition: l968;47 :10-22.
1. Those caused by raised intra-articular joint Eriksson E. Rehabilitation of muscle function after sport injury : major
pressure (de Andrade et al. 1965). problem in sports medicine. mt J Sports Med 1981 ;2:l-6.
Eyring EJ, Murray WR. The effect ofjoint position on the pressure of
2. Other impulses, non-pressure-mediated, shown
intra-articular effusion. J Bone and Joint Surg [Am]
by a further increment of strength after intra-articular 1964;46-A:1235-41.
anaesthesia. It is not clear whether these afferents are Gerber NJ, Dixon AStJ. Synovial cysts and juxta-articular bone cysts
(Geodes). Semin Arthritis Rheum l974;3 :323-48.
related to the pain of inflammation.
Harding AEB. An investigation into the cause of arthritic muscular
Both mechanisms seem to work through other than
atrophy. Lancet 1929:1:433-4.
ordinary pain reflexes, since none of our patients had Jayson MIV, Dixon AStJ. Intra-articular pressure in rheumatoid
pain on muscle contraction before aspiration or anaes- arthritis of the knee. I. Pressure changes during passive joint
distension. Ann Rheum Dis l970;29:26l-5.
thesia (Dee 1978; Eriksson 1981 ; Newton 1982; Wood
and Ferrel 1984). This has been confirmed by studies on Jones DW, Jones DA, Newham Di. Chronic knee effusions and
aspiration : the effect on quadriceps inhibition. Br J Rheumato/
the reflex inhibition of quadriceps activity after menis- 1987 ;26 :370-4.
cectomy and arthrotomy (Stokes and Young 1984; Kennedy JC, Alexander IJ, Hayes KC. Nerve supply ofthe human knee
and its functional importance. Am J Sports Med 1982;l0:329-35.
Shakespeare et al. 1985).
Levick JR. Joint pressure-volume studies : their importance, design and
Our results correlate well with the earlier findings of
interpretation. J Rheumato/ 1983 ;lO :353-7.
Spencer et al. (1984) in healthy volunteers when
Lieb FJ, Perry J. Quadriceps function : an anatomical and mechanical
simulated effusions of comparable volume (60 ml) study using amputated limbs. J Bone Joint Surg [Am]
1968 ;50-A :1S35-48.
inhibited the H-Reflex by 31 to 44%, corresponding to an
estimated 8.5% of the total pool of motor neurones. Machan FG. Die Rolle des intraartikul#{228}ren Druckes groser Gelenke
am Menschen und seine klinische Relevanz. Z Arzt/ Fortbi/d
Unlike ours, their measurements were based on a neuro- (Jena) 1983 ;77 :877-81.
physiological reflex which was independent of conscious McCarty DJ. The physiology ofthe normal synovium : I. Intra-articular
pressure. In : Sokoloff L, ed. The joints and synovia/ fluid. New
control. Our measurements depended entirely on our
York, etc : Academic Press, 1980:294-6.
patients’ motivation and the voluntary contraction of
Moritani T, deVries HA. Re-examination of the relationship between
muscle. the surface integrated electromyogram (IEMG) and force of
isometric contraction. Am J Phys Med 1978;57:263-77.
The practical consequence of our observations is
Newton RA. Joint receptor contributions to reflexive and kinesthetic
that the joint should be aspirated and systemic or intra-
responses. Phys Ther 1982;62:22-9.
articular anti-inflammatory drugs be given before start- Shakespeare DT, Stokes M, Sherman KP, Young A. Reflex inhibition of
ing quadriceps exercises if maximal training effects are the quadriceps after meniscectomy : lack of association with pain.
C/in Physio 1985;5:137-44.
to be achieved.
Soderberg GL, Cook TM. An electromyographic analysis of quadriceps
We wish to thank Dr A. StJ. Dixon (Bath) for critically reading and femoris muscle setting and straight leg raising. Phys Ther
improving the manuscript. 1983 ;63 :1434-8.
No benefits in any form have been received or will be received Spencer JD, Hayes KC, Alexander IJ. Knee joint effusion and
from a commercial party related directly or indirectly to the subject of quadriceps reflex inhibition in man. Arch Phys Med Rehabil
this article. l984;65 :171-7.
Staudte HW. Die innere Mechanik der Muskulatur, theoretische
REFERENCES Untersuchungen normaler und atrophischer Muskeln. Z Orthop
1981 ;l 19:266-76.
Andrews JG. On the relationship between resultant joint torques and Stokes M, Young A. The contribution of reflex inhibition to
muscular activity. Med Sci Sports Exerc 1982;14:361-7. arthrogenous muscle weakness. C/in Sci 1984;67:7-14.
Basmajian JV. Re-education of vastus medialis : a misconception. Arch Stratford P. Electromyography of the quadriceps femoris muscles in
Phys Med l970;51 :245-7. subjects with normal knees and acutely effused knees. Phys Ther
1981 ;62:279-83.
Bittscheidt W, Hofmann P, Schumpe G. Elektromyographische
Untersuchung an der Oberschenkel-muskulatur bei h#{228}mophilem Wood L, Ferrell WR. Response of slowly adapting articular mechanor-
Kniegelenkserguss und bei Reizzust#{228}nden des Kniegelenkes. Z eceptors in the cat knee joint to alterations in intra-articular
Orthop l978;1l6:56-60. (Eng. Abstr.) volume. Ann Rheum Dis 1984;43:327-32.

THE JOURNAL OF BONE AND JOINT SURGERY