THE EFFECTS OF ACTIVE AND PASSIVE NECK FLEXION ON UNILATERAL STRAIGHT-LEG-RAISING Craig Smith BSc Physiotherapy* INTRODUCTION The straight-leg-raise (SLR) test has been previously reviewed with respect to standardising the protocol and high­ lighting relevant aspects of the literature1. SLR tests the integrity of the spinal neural structures, namely the spinal cord, nerve roots and dura mater and serves as an objective diagnostic test for lumbar spinal pathology. One of the qualifying tests associated with SLR is neck flexion (NF). The application of NF is such that increased tension is applied to these neural structures when performed either with SLR 2-8 or independently from it9, in order to elicit any obstruction to their free movement in the intervertebral canal or foramen. As the angles that the emergent thoracic and lumbar roots make with the cord are smaller than those made by the cervical nerve roots, there exists greater resistance to cord tension originating cranially rather than caudally10. NF pro­ voking sciatic symptoms will therefore imply an increase in tension in the root sheaths and dura mater due to an increase in the bending tensile forces caused at, for example a disc lesion site - known as Hyndman’s Sign 10,11 or Brudzinski’s T e st3’11’12’13. The bulk of the literature documents that NF be applied passively when added to the SLR test3’4,5’12’14’15. Clinically, it proves quite cumbersome to try and maintain the SLR with one hand while passively flexing the neck with the other. Documentation of active neck flexion (ANF) is minimal and non-specific. Cailliet16 and Cyriax17 note it to be per­ formed by the patient. Cyriax and Cyriax18 demonstrate it pictorially while Hoppenfeld19 mentions that in Kernig’s Test, the patient forcibly flexes the neck. None of the authors specify whether NF needs to be applied passively or actively. Surely there can be no difference to the SLR test, as with both methods, tension is applied to the dura and cord and thus transmitted to the lumbosacral roots? Personal experience and communication with col­ leagues have brought to light some discrepancy between the amount of SLR available when applying passive neck flexion (PNF) and then ANF on normal subjects (unpublished obser­ vations). When ANF was performed, a definite reduction in the resistance of SLR at the end of range (EOR) was ob­ served. The thigh could be flexed a few degrees further on the hip, and the new EOR noted when resistance to the movement was again felt. No concrete, scientific explanation could be attributed to this phenomenon, and thus it was felt necessary to explore SUMMARY A study was designed to test the effects of passive neck flexion (PNF) as compared with active neck flex­ ion (ANF) on the straight-leg-raise (SLR) test. Twenty two subjects were measured three times with a goniometer and the difference in range of movement (ROM) was determined by subtracting the SLR recor­ dings obtained by means of PNF from those of ANF. Results demonstrated that neck flexion (NF) performed actively allowed a mean value of 4,13° more SLR than when performed passively. The implications of these findings are discussed. OPSOMMING ’n Studie is ontwerp om die uitwerking van passiewe nekfleksie (PNF) teenoor aktiewe nekfleksie (ANF) op die reguitbeen-oplig toets (RBO) te ondersoek. Bogenoemde ondersoeke is op twee-en-twintig per- sone met behulp van 'n goniometer drie keer gemeet, en die verskil in bewegingsomvang is bereken deur die RBO resultate behaal deur middel van PNF van die van ANF af te trek. Resultate het gewys dat ANF 4,13° meer RBO toege- laat het as die van PNF. Die implikasies van die bevin- dinge is bespreek. further by formulating an experiment testing the effects of both ANF and PNF on SLR, in order to determine whether there exists any difference in the amount of SLR range when both are individually and successively superimposed. If ANF does allow significantly more range of SLR movement than PNF, then its implications for the test would be such that, with more range of motion available, more tension could be exerted upon the nerve roots and thus the quality o f symptoms produced could be improved. METHOD 1. SUBJECTS Twenty two healthy male (12) and female (10) students with a mean age of 20,8 years (SD = 1,65) were tested at the UCT Physiotherapy Department. None of those tested had any known orthopaedic or neurologic dysfunction. Partici­ pants were instructed not to begin any new stretching exer­ cises or alter their physical activities during the study period. 2. GONIOMETRIC VALIDATION The reliability and repeatability for the intratester go- niometric measuring was evaluated prior to the SLR testing procedure. Firstly, a female’s right leg was immobilised in hip flexion (same surface markings described in procedure), and * Currently studying BSc (Med) (Hons) Sports Science, UCT. This study was submitted in partial fulfillment of the requirements for BSc Physiotherapy at the University of Cape Town). Physiotherapy, February 1990, vol 46 no 1 Page 11 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) seven measurements were taken by the tester with the same goniometer. The same leg was then immobilised at a fixed angle in the SLR position using the pulley system to be described, and a further seven measurements were taken as before. These measurements were then analysed for vari­ ability (coefficient of variation) to determine intratester relia­ bility and repeatability and are reported in the results section. 3 . INSTRUMENTATION An overhead pulley system was used to maintain the passive SLR on each subject’s right leg. The range of move­ ment (ROM) was measured with a plastic long arm go­ niometer, which was found to be of high reliability when compared with the tape measure and flexometer for recor­ ding SLR20. A pressure transducer (calibrated in millimetres mercury [mmHG]) was used to indicate the EOR by measur­ ing the resistance o f the hamstrings and posterior thigh struc­ tures. This method was employed as subjective estimates of EOR would have varied considerably due to different pain threshold levels between subjects. A plasterzote knee backslab was applied to each right leg with a crepe bandage in order to prevent knee flexion. A sling was placed beneath each subjects’ right heel, and pro­ vided the point of attachment for the pulley’s rope, as well as keeping the pressure transducer in place. The rope was then passed through a hook above the subject’s right hip so that its force was exerted at about 90° to the leg when SLR was performed. 4 . PROCEDURE In order to measure the angles of SLR, the following anatomical landmarks were delineated: - a) the greater trochanter b) a point on the line connecting the greater trochanter and the lateral femoral condyle c) a point on the horizontal line from the greater tro­ chanter to the mid axilla. After being marked, each subject was then positioned on the plinth in supine and the backslab applied. The pressure transducer was preset for each measurement at 40 mmHg, and placed under the heel within the sling. The rope was connected to the sling and passed through the hook, and the free end given to the subject. They were then instructed to relax their hamstring muscles, and not to resist or assist the SLR in any way. Figure 1: SLR starting position showing the knee backslab, the pulley and the sling Each subject’s right leg was then raised by the tester, (with attention to the SLR biomechanical detail as described before1) until the pressure gauge value read 120 mmHg. At this stage each subject was asked to hold the opposite end of the rope with both hands while keeping their elbows on the plinth. This maintained the SLR position and enabled the tester to read the pressure values and measure the SLR range with the goniometer. Figure 2: The subject stabilising the leg once passive SLR has been performed PNF was then applied, and the gauge consulted to see if any changes in pressure had occurred. If so, these were recorded. The head was then lowered, and the gauge re­ checked to ensure it read 120 mmHg. The subject was then instructed to actively lift his head himself, approximating his chin to his chest while keeping his shoulders on the plinth. The pressure gauge was again checked and the changes (de­ creases) in values were then recorded. With the neck still flexed, the leg was then raised further by the tester until the- gauge read 120 mmHg again. This new range of SLR was then measured and recorded. Only then was the subject told to lower his head and leg to the plinth. Figure 3: ANF superimposed on passive SLR. This procedure was performed three times on each subject, with a one minute rest in between. The data was analysed for statistical significance using the student’s t-test. RESULTS The coefficients of variation for the goniometric meas­ urements o f hip flexion and SLR were 1,5% and 1,3% respec­ tively. Values less than 10% are regarded as acceptable, and Bladsy 12 Fisioterapie, Februarie 1990, deel 46 no 1 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) Q CD O) i- CD QJ thus the above results suggest highly acceptable intratester reliability and repeatability, with the possibility of goniome- tric error being low. There were no changes in pressure readings when PNF was applied to SLR, while with ANF, the readings decreased by a mean value of 6,5 mmHg. SLR. Procedure no. n rnear^ s e m t-value DF P 1 22 3.91 0.48 8.07 21 0.001 2 22 3.91 0.46 8.41 21 0.001 3 22 4.59 0.46 9.88 21 0.001 Average 22 4.13 0.47 8.79 21 0.001 TABLE: Mean, standard error mean, t-value. degrees of freedom and probability for the increased range in degrees of SLR with ANF From the table, it can be seen that the angle of SLR was greater with ANF than with PNF during all three procedures. These values correspond to a significant average mean in­ crease of 4.13° (p < 0.001). These increases were calculated by subtracting the SLR ranges with PNF from those with ANF. Figure 4 depicts the relationships between the three procedures’ mean initial and final measurements. For proce­ dures one and two, both initial and final readings correlate very closely, varying by only 0,24° and 0,23° respectively. The third procedure’s readings were greater than those of the first two, and varied from their combined mean average for the initial and final readings by 1,73° and 2,31°. These increases can be attributed to the possible stretch­ ing and relaxing of the hamstring muscles by the time the third measurements were taken. This might have indicated that some form of stretching exercises should have been per­ formed before measurements were taken, to minimise the hamstrings from stretching further during the three proce­ dures. DISCUSSION The results of this study clearly indicate that more SLR was achieved with A N F than with PNF. Analytical compari­ sons showed a statistically significant difference between the two ( p < 0,001), and it is thus evident that the method of applying NF does affect the range of SLR. The intratester reliability and repeatability of the go- niometric measurements were well supported by the coeffi­ cients of variation for hip flexion (1,5%) and for SLR in the sling (1,3%), and compare with reports by Hsieh et al20 that the goniometer is of high intrasession and intersession relia­ bility and repeatability. What then are the causes for ANF allowing more SLR then PNF? The most likely answer to this question involves the abdominal muscles. When in supine, flexing the neck is an action against gravity, and is performed by the two sternoclei­ domastoid muscles working in unison. With this movement being performed, overflow to the abdominal muscles occurs. This can be demonstrated by lying on your back and placing your hands on your abdomen. When you lift your head off the surface, the abdominals are immediately felt to contract tight­ ly. They originate from the sternum and lower ribs and insert into the iliac crests, the superior pubic rami and the symphysis pubis. Thus, when they contract, tension results between the two points of fixation and if either point is not stabilised, movement can occur . In this case, the chest wall and ster­ num are rigid, and thus the movement takes place at the pelvis. The abdominal insertions will cause the anterior pelvis to rotate forwards and upwards, and because the back extensors which insert into the posterior pelvis are relaxed, it will rotate backwards and down­ wards. This pelvic rotation thus relaxes the tension in the hamstring muscles and posterior thigh struc­ tures, because of their insertions to the postero-infe- rior aspect of the pelvis. Posterior pelvic rotation occurring with SLR begins as early as 10° of hip flexion23, and can be referred to as the pelvifcmoral motion”2'25. Results have shown that at a maximum SLR of 87,3°, contri­ buting factors were increases of 32,1° in the pel­ vis/horizontal angle, and 55,2° in the SLR/pelvis angle23. Thus, if this much pelvic rotation were oc­ curring just with SLR motion, it would indicate that the pelvis is not completely stabilised during SLR, and applying the force of the abdominals to it during ANF, could result in further rotational movement. This pelvic rotation thus exerts its effects on the SLR leg through the pelvifemoral motion. The increased posterior pelvic rotation at the end of SLR slightly relieves the tension in the hamstrings, and thus more range of SLR is available, until the hamstrings are once again tight and resisting the movement. In this study, the increase in pelvic rotation was found to be 4,13°, (from 83,10° to 87,23°) and relates directly to the hypothesis that more SLR can be achieved when ANF rather than PNF is superimposed. CLINICAL IMPLICATIONS If this is then the case, how do these findings relate to the clinical environment. Firstly, applying NF to the SLR\test increases the tension in the dura mater and the lumbosacral nerve roots. Whether applied passively or actively, this resultant tension is achieved. However, A N F applied to SLR adds a few more degrees to the SLR angle. This should then apply further tension to 9 4 9 2 9 0 88 86 8 4 82 P ro ce d u re s In itia l SLR IHH Final SLR w i t h ANF Figure 4: Mean and sem values for initial and final SLR measurements Physiotherapy, February 1990, vol 46 no 1 Page 13 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) the sciatic nerve and its roots, making the test more specific. As previously documented1, the root tension and move­ ment with SLR is considered to increase only up to 70°, after which it becomes sluggish and ceases at 90 ’ . It would then seem that ANF used with SLR might only be effective if the end o f the SLR angle is anywhere less than or equal to roughly 70°. If SLR is limited by back and/or leg pain comparable to sciatic nerve involvement, then the purpose of the NF qual­ ifying test is only to reproduce or enhance this pain. No further range o f SLR is required (as this would only exacer­ bate the patient’s pain and make it quite unbearable) and thus ANF or PNF can be used. Because we know that SLR also causes lumbo-pelvic movement, irritable facet or sacroiliac joints can also produce comparable back pain1. Therefore, NF could be used as a differentiating test for neural or articular structure involve­ ment1. Pain on further SLR reproduced with ANF can either be due to stretching of the dura mater or movement of the irritable lumbosacral joints, due to pelvic rotation caused by the abdominals. If PNF is then applied and the pain enhanced, then this can only be attributed to the fact that the dura and nerve roots’ mobility are being painfully impaired, for there does not exist any anatomical means by which PNF can exert any effect on the sacroiliac or facet joints, or even be thought o f to be a cause for pelvic movement. If PNF does not cause an increase in painful symptoms, then the cause can most likely be attributed to the irritable lumbo-pelvic joints which are being mobilised with SLR and ANF, or neural structures unaffected by NF (e.g. disc pro­ lapse situated inferomedially to the nerve root and dura mater28). If ANF exerts an effect on the pelvis and indirectly on the hamstrings, are there any other structures which might also be affected? The isometric contraction o f the abdominal musculature with ANF helps to decrease intradiscal pressures29 by in­ creasing intra-abdominal pressure. The increase in intra-ab­ dominal pressure together with the splinting effect o f the abdominal muscles, reduces the pressure on the discs. Fur­ ther studies would be needed to determine if there exists any abating of discal symptoms with the decrease in intradiscal pressure when ANF is applied. Fisk30’31, referring to hamstring tightness and SLR, notes that “limited SLR by the hamstring muscles can be improved with manipulation of the lumbar spine, resulting in either alteration of the gamma inflow to the muscle spindles allowing lengthening o f the hamstrings, or sacro-iliac joint mobilisation”. That ANF might exert an effect on the ham­ strings by also altering the gamma inflow to the muscle spind­ les or any other neurological pathway can only be seen as speculation, and further insight and experimentation would be needed to determine if any relationship does exist. RECOMMENDATIONS Having analysed and tabulated the results for the twenty- two normal subjects who took part in this experiment, it can be seen that it can only be regarded as a pilot study. The study itself needs to include a larger number of candidates, span­ ning a wider age group; 12-16 years and 35-55 years. More care also needs to be taken to attempt to control the interfering variables which might have affected the results. These were; 1. movement of the skin where the bodymarkings were delineated, such that they were no longer over the bony landmarks being represented, when SLR measurements were taken ' 2. leg weight differences between subjects, exerting varied forces on the pressure transducer 3. the position of the hook o f the pulley above the right hip. Visual estimates o f the point above the right hip were used, and varied positions could have caused different angles of pull o f the rope on the SLR leg. Further research involving ANF, pelvic rotation and SLR would be beneficial in helping to understand this phe­ nomenon. 1. Tests performed in the clinical environment involving subjects with back symptoms, to determine whether A N F and PNF can actually be used in a differentiating role, for patients who exhibit articular and neural symp­ toms. 2. A study to determine how much more pelvic rotation actually does occur when ANF is added to full SLR. 3. A N F tested on passive and active SLR. 4. Lastly, experimentation to determine the neurological effects of ANF on hamstring muscle spindle control. CONCLUSION The angle of SLR with A N F was definitely more than the angle with PNF. This bears important clinical relevance with regard to how NF is performed, as well as qualifying the signs produced with SLR. Emphasis is not placed on demanding the sole use of ANF, but with the clinician’s discretion, it can be useful in:- i) applying more tension to the nerve roots, and ii) differentiating between neural and articular low back pain causes. REFERENCES 1. Smith CA. Analytical Literature Review of the passive Straight-Leg-Raise Test. South African Journal o f Physiotherapy 1989;45(4):104-107. 2. Cyriax J. Dural Pain. Lancet 1978;1:919-921. 3. Grieve GP. Mobilisation o f the spine. 4th ed. Edinburgh : Churchill Livingstone, 1976:46-48. 4. Maitland GD. Vertebral Manipulation. 5th ed London : Butterworth, 1986:67-71. 5. Breig A, Troup JD G . Biomechanical considerations in the straight-leg- raising test. Spine 1979;40:242-250. 6. Edgelow PI. Physical examination o f the lumbosacral complex. P/tys Ther 1979;59(8):974-977. 7. Breig A, Marrions O. Biomechanics of the lumbosacral nerve roots. Acta Radiol 196;1:1141-1160. 8. Reid JD . Flexion extension movements o f the head. J Neurol Neurosurg Psychiatry 1960;23(3)214-221. 9. Edgar MA, Park WM. Induced pain patterns on passive straight-leg-rais- ing in lower lumbar disc protrusions. JB one Joint Surg 1974;56B:658-667. 10. Breig A. Adverse Mechanical Tension in the Central Nervous System Stockholm : Almqvist and Wiksell, 1978:14-15,38-43,152-170. 11. U rban LM. SLR t e s t : a review. In : Grieve GP, ed. M odem Manual Therapy o f the Vertebral Column. Edinburgh: Churchill Livingstone, 1986:567-575. 12. Brody IA, Wilkins RH. The signs of Kemig and Brudzinsky.^rcA Neurol 1969;21:215-216. 13. Wilkins RM, Brody IA. Lasegue’s Sign./4/cA Neurol 1969;21:219-220. 14. Scham S, TaylorT. Tension signs in the lumbar disc prolapse. Clin Orthop 1971;75:195-204. - 15. Jull GA. Examination of the lumbar spine. In: Grieve GP, ed. M odem Bladsy 14 Fisioterapie, Februarie 1990, dee! 46 no 1 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) Manual Therapy o f the Vertebral Column. Edinburgh: Churchill Living­ stone, 1986:556-557. 16. Cailliet R. Low Back Pain. 3rd Ed. Philadelphia : F A Davis Company, 1981; 152-158,178-180. 17. Cyriax JH . Textbook o f Orthopaedic Medicine Vol l 7th ed. London : Balliere Tindall, 1982 : 378-389,414-119. 18. Cyriax JH, Cyriax PJ. Illustrated Manual o f Orthopaedic Medicine. Lon­ don: Butterworths, 1983:188. 19. Hoppenfeld S. Physical Examination o f the Spine and Extremities. New Y o rk : Appleton Centuiy-Crofts, 1976:256-259. 20. Hsieh C, Walker JM, Gillis K. Straight-leg-raising test: Comparison of three instruments. Phys Ther 1983;63(9): 1429-1433. 21. Galley PM, Forster A L Human Movement. Edinburgh: Churchill Liv­ ingstone, 1982:32. 22. Bohannon RW. Cinematographic analysis of the passive straight-leg-rais­ ing test for hamstring muscle length. Phys Ther 1982;62(9):1269-1274. 23. Bohannon RW, Gajdosik RL, Le Veau BF. Contribution of pelvic and lower limb motion to increases in the angle of passive straight-leg-raising. Phys Ther 1985;65(4):474-»76. 24. Gajdosik RL, Le Veau BF, Bohannon RW. Effects of ankle dorsiflexion on active and passive u n ilateral straight-leg-raising. Phys Ther 1985;65(10): 1478-1482. 25. Bohannon RW, Gajdosik RL, Le Veau RF. Relationship of pelvic and thigh motions during unilateral and bilateral hip flexion. Phys Ther 1985;65(10): 1501-1504. 26. Chamley J. Orthopaedic signs in the diagnosis of low back pain. Lancet 1951;1:186-189. 27. Fahmi WM. Observations on straight-leg-raising with special reference to nerve root adhesions. C anJSurg 1966;9(l):44-48. 28. Kessler RM. Acute symptomatic disc prolapse: Clinical manifestations and therapeutic considerations. Phys Ther 1979;59(8):978-987. 29. Jensen GM. Biomechanics o f the lumbar intervertebral disc: A Review. Phys Titer 1980;60(6):765-773. 30. Fisk JW. The passive hamstring test: Clinical evaluation. N A M ed J 1979;89(631):209-211. 31. Fisk JW. The straight-leg-raising test: its relevance to possible disc pathology. N Z M ed J 1975;81(541):557-560. * CLASSIFIED ADVERTISEMENTS URGENTLY REQUIRED Part-time Physiotherapist experienced in in-patient orthopaedics (mainly trauma) to assist in clinical supervision/teaching of third and fourth-year physiotherapy students. Hours: 3 mornings per week, 08:00 12:00 University holidays free. Salary: 3/8 of appropriate notch of Junior Lecturer's scale, according to experience. Starting date: 1 February 1990, or as soon as possible thereafter. APPLICATIONS should be submitted to the following: Miss S Irwin-Carruthers, Department of Physiotherapy P O Box 63, Tygerberg 7505. PRETORIA Full time post available in orthopaedic practice. Conditions treated include joint replacements and rehabilitation following specific shoulder surgery. CONTACT: Carol Castleman, (012) 343-1298 (w), 57-6373 (h). ELIM HOSPITAL, Northern Transvaal. Physiotherapist urgently required for hospital and community work (TPA post). CONTACT: Dr Jacques (Medical Superintendent) P O Box 12, Elim Hospital, 0960, or telephone (01552) 611, exten­ sion 126 or 216. PAARL A community based project in Mbekweni (Paarl) would like to em­ ploy a Physiotherapist on a sessional basis to work with stroke pa­ tients. The work will involve assessment of patients as well as the training and supervision of a community based health worker. Experience in adult neurology is an advantage but not essential. Ability to speak Af­ rikaans and/or Xhosa will also be an advantage. Transport will be available from Cape Town. PHONE: Ian MacKenzie at (021) 47-7863 (oh) or 47-9073 (ah), or apply to P O Box 234, Salt River 7925. CLASSIFIED ADVERTISEMENTS CAPE TOWN Full time physiotherapist required for a private practice in central Cape Town. Work involves cardio-thoracic intensive care and in- and out- chest patients. Starting April - July 1990. PLEASE CONTACT: Mrs S Pollock, page (021) 61-634, or 531-8619 (ah). PRACTICE AVAILABLE IN ESSEX Busy Essex Practice for sale. Fully equipped including receptionist and staff. Close to local schools; bus link, shops and railway. Referrals from local GP's and Orthopaedic Specialists in the area. Well established over the past six years, currently seeing a 100+ pa­ tients per week and enjoying an excellent variety of spinal and pe­ ripheral/orthopaedic conditions. London within 55 minutes by train or car. WRITE TO: LRT Brown, 49 High Road, Rayleigh, Essex SS6 7SE, or telephone 0268-777641. ASSOCIATION FOR CHILDBIRTH AND PARENTHOOD 5th Workshop for professionals who deal in childbirth and parent- hood. Date: Saturday 24 March 1990 Time: 10h30 15h30 Venue: University of Pietermaritzburg Topics: “Teenage Pregnancy" “Open Adoption” Fee: (Includes Lunch) Non-members R35,00 ACP Members R25.00 Students R15.00 To book contact: Doreen O’Neill, P O Box 22491, Glenashley 4022, Tel: (031) 52-5128. NEW PRACTICE IN HILLBROW Arie Michaeli wishes to inform his colleagues that as from 1st January, 1990 he will be practicing privately at: 416 Tower Hill, Kotze Street, Hillbrow, Johannesburg. Office hours: (011) 724-2359, after hours: 643-6473. NOTICE OF MEETING The Neuro-rehabilitation Group will next meet on Saturday 17 Fe­ bruary 1990, in the Physiotherapy Department, Johannesburg Hospi­ tal, at 9.00 am. Subjects for discussion: (1) Cognitive changes in head injuries (2) Rhizotomies. Physiotherapist SETTLERS HOSPITAL, GRAHAMSTOWN • Duties: Administering of Physiotherapy to ^-/out­ patients at the abovementioned hospital. • Registration with the S.A. Medical and Dental Council as a Physiotherapist is essential. • Salary: A salary to a maximum of R25 371 per annum may be granted according to qualifications and experience. Application must be made on the prescribed form (Z.83, obtainable from any government office) and forwarded to the Medical Superintendent, Settlers Hospital, Private Bag 1007, Grahamstown 6140. < v Closing date: 28 February 1990. LINTAS i 5879 Physiotherapy, February 1990, vol 46 no 1 Page 15 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. )