R e s e a r c h A r t i c l e T h e R o l e o f P r o p h y l a c t i c C h e s t P h y s i o t h e r a p y A f t e r C a r d i a c V a l v u l a r S u r g e r y : Is T h e r e O n e ? This work was approved by the University of the W itwatersrand in partial fulfilment of the degree of M aster of Science in Physiotherapy. A B S T R A C T : A study was conducted to investigate whether there is a role f o r prophylactic chest physiotherapy in:preventing pulm onary complications in-patients undergoing uncomplicated cardiac valvular surgery. 'Thirty patients were randomly divided into two groups. The patients in group I (n = 14) received instructions that they should cough after their operation and that they should get out o f bed on the second post-operative day. The patients in group 2 (n = 16) were seen pre-operatively and treated post-operatively by a physiotherapist. The treatment consisted o f breathing exercises, supported coughing and assistance with walking fro m the second post-operative day. These patients were seen twice daily on the fir s t and second post-operative days and once daily on the third and fo u rth post-operative days. No patients in either group developed a pulm onary complication during their post-operative hospital stay. The incidence o f hypoxaemia, chest X-ray abnormalities and length o f post-operative stay were not signi­ fica n tly higher f o r the control group. The findings o f this study suggest that the role o f prophylactic chest physio­ therapy follow ing cardiac valvular surgery is questionable. KEYWORDS: PH YSICAL THERAPY, PHYSIOTHERAPY, BREATHING EXERCISES, OPEN H E A R T SU RG ERY DE CHARMOY SB, MSc'; EALES CJ, PhD 1 Department of Physiotherapy, University of the W itw atersrand INTRODUCTION Physiotherapists working in the cardio- thoracic wards spend a large proportion o f time each day assessing and treating post-operative cardiac patients. In the last two decades, much research into differ­ ent com binations o f chest physiotherapy treatm ent techniques has occurred in an attempt to establish the most effective com bination in preventing post-oper- ative pulmonary com plications. Recent research has com pared specific chest physiotherapy regimes to a program m e of walking the patients as soon as possi­ ble post-operatively (Jenkins et al, 1989; Jenkins et al, 1990; Stiller et al, 1994; Stiller et al, 1995). The research by Jenkins and Stiller has been largely confined to patients undergoing coronary CORRESPONDENCE: Sue de Charmoy Physiotherapy Department University o f the W itwatersrand 7 York Road Parktown 2193 Johannesburg South Africa Tel: (27) 11 488-3450 Fax: (27) 11 488-3450 Email: 159suz@ chiron.wits.ac.za artery bypass graft surgery. These resear­ chers have consistently reported that pro­ phylactic physiotherapy in this patient population is o f little value in preventing po st-o p erativ e pulm onary co m p lic a ­ tions. It must be em phasised at this point that the patients in all these studies were either mobilised by the nursing staff or given an instruction to mobilise. Jenkins et al (1994) reported the inci­ dence o f respiratory com plications post- operatively as 9% despite the lack of chest physiotherapy. THe patients excluded from this study included those who had had previous coronary artery surgery or pulm onary surgery, and those that had a pre-operative respiratory abnormality. In 1995 Stiller eta l. investigated whether the incidence o f clinically significant pulmonary com plications had increased since the recom m endation that routine post-operative physiotherapy was not necessary in the uncomplicated coronary artery surgery patient. The 1995 study included all patients undergoing heart sur­ gery requiring cardiopulm onary bypass. The only difference in this study from the 1994 study was that it included 13 patients who had undergone cardiac valve surgery w ithout coronary artery surgery. Clinically significant pulmonary com plications were found in 7.1% (nine out o f 127 patients) o f the total patient population. An im portant consideration in this study is that all patients under­ going cardiac surgery were included and thus patients with significant pre-oper- ative risk factors w ere also included. A study by Johnson et al (1996) included 75 patients who had undergone ,valve surgery. The mean age o f the patients in this study group was 63 ± 1 2 and 68 ± 10 years.. The patients were random ly divided into two groups. One group received a lower intensity treatment w hich included education, early am bula­ tion, and deep breathing exercises, while the other group received the sam e treat­ m ent with the addition o f single handed percussion (higher intensity treatment). They reported a 5% respiratory com pli­ cation rate. For the purposes o f their study a respiratory com plication (pneumonia) was defined as the presence o f three out o f four of the following variables: a white blood cell count greater than 109/L; anO oral tem perature, greater than 38.5 C; a p o sitiv e cu ltu re fo r a resp irato ry pathogen in the sputum and evidence of air bronchogram s on chest X-ray. 24 S A Jo u r n a l o f Ph ysiotherapy 2000 V o l 56 No 3 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. ) mailto:159suz@chiron.wits.ac.za Table 1. Individual patient data fo r patients in the gro up 1. Patient Age Sex Height (m) W eight (Kg) BMI (Kg/m 2) 1 29 F 1.54 67 .3 0 28.38 7 50 F 1.54 5 7 .0 0 24.03 9 11 F 1.26 22.50 14.17 11 26 F 1.59 50 .2 0 19.86 13 12 F 1.39 30.10 15.58 15 35 F 1.58 6 7 :8 0 27.26 16 38 F 1.55 8 0 .5 0 33.51 19 25 F 1.64 6 5 .3 0 24.28 20 22 F 1.44 5 0 .2 0 24.21 21 29 F 1.56 4 3 .0 0 17.67 28 40 F 1.51 4 9 .8 0 2 1 .9 9 29 15 F 1.54 46 .2 0 19.48 31 30 M 1.61 5 3 .8 0 20.76 32 36 F 1.56 5 4 .7 0 22.48 Mean 28.43 1.52 52 .7 4 22.4 *S.D ± 1 1 .1 2 ± 0 .0 9 ± 0 .0 9 ±5.16 * S.D. = Standard deviation Table 2. Individual patient data fo r patients in gro up 2. Patient Age Sex Height (m) W eight (Kg) BMI (Kg/m 2) 3 16 F 1.28 39 .7 0 24.23 4 13 M 1.38 29 .1 0 15.28 5 44 F 1.60 84 .2 0 3 2 .8 9 6 11 M 1.26 22 .0 0 13.86 8 53 F 1.60 5 0 .3 0 19.65 12 42 F 1.67 70 .7 0 2 5 .3 5 17 23 F 1.63 52 .2 0 19.65 18 12. M 1.34 25.60 14.26 22 32 F 1.55 4 4 .1 0 18.36 23 4 0 M 1.54 4 8 .0 0 2 0 .2 4 25 63 M 1.84 85.60 25 .28 26 23 M 1.76 67.40 21 .7 6 27 4 0 F 1.60 4 8 .3 0 18.87 30 33 M 1.64 61 .9 0 23.16 33 24 F 1.58 4 4 .2 0 17.71 34 27 M 1.72 56 .0 0 18.93 M ean 31 1.56 51.83 2 0 .5 9 *S.D. ±15.21 ±0.17 ± 1 8 .7 9 ± 4.84 * * S.D. = Standard deviation In South A frica the p o pulation encountered who undergo valve surgery is m arkedly different to that o f the first world due to the prevalence o f rheumatic heart disease. In South Africa a study conducted in 1972 by M cLaren et al. surveyed 12 050 black schoolchildren in Soweto. They found an overall preva­ lence rate of rheum atic heart disease of 6.9 per 1000 with a peak rate o f 19.2 per 1000 in children aged 15 to 18 years (M cLaren et al, 1975). The incidence of rheum atic heart disease has been on the decline in the developed world for many years. The decline is evident in the fol­ low ing statistics for the U nited States of America. In 1950 the death rate per 100 000 was 14 and in 1993 it was 1.7 per 100 000 (Am erican H eart Associa­ tion, 1997). In South Africa patients with rheu­ matic valve disease, especially those p resen tin g w ith active card itis, are young, black and have relatively under­ privileged backgrounds (M cLaren et al, 1975; Barlow 1992a and 1992b; M arcus et al, 1994). In the year 1 June 1994 to 31 M ay 1995, 313 patients underwent cardiac valve surgery at the Johannes­ burg hospital. There were 239 black, 27 w hite, 14 asian and 33 patients o f m ixed racial descent. O f these 182 were female and 131 male. Seventy-six of these patients were under the age of 15. In contrast, during the same period, 118 patients underw ent coronary artery surgery, thus indicating the need for research involving these valvular surgery patients. It was bearing this in mind that a study was conducted to determ ine if the results obtained by Stiller et al (1994 and 1995), Jenkins et al (1989 and 1994) and Johnson et al (1996), could be app lied to the South A frican valve surgery population. METHODS The research hypothesis for this study was that routine post-operative physio­ therapy, which includes a regimen of breathing exercises, coughing and w alk­ ing, is o f no benefit in the uncomplicated post-operative valvular surgery patient. Subjects All patients adm itted for elective cardiac valvular surgery over a three-m onth period who signed inform ed consent were considered for inclusion in the study. A total of 36 patients out of 38 adm issions were included. Patients excluded pre-operatively w ere those ' with a docum ented history o f pulmonary disease or a neurological disorder which interfered with their ability to partici­ pate in the study. Two patients were excluded from the study pre-operatively as a result o f a docum ented history of pulm onary tuberculosis. Post-operatively patients were withdrawn from the study if they w ere intubated for a period longer than 24 hours, had an intensive care stay of longer than 48 hours or returned from theatre with a neurolo­ gical or cardiac com plication w hich rendered them unstable or unable to par­ ticipate in the study. Six patients were w ithdrawn from the study during the post-operative period. Three o f the patients suffered a cerebro­ vascular accident during surgery or in the SA Jo u r n a l o f Physiotherapy 2000 V o l 56 No 3 25 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. ) im m ediate post-operative period, one patient was intubated for longer than 24 hours and the other two patients remained in the intensive care unit for longer than 48 hours due to cardiac instability. The total study population was 30 patients o f w hich 21 were fem ales and nine males. The mean age of the group was 29.72 years. Groups Participants were allocated a num ber in order o f admission. These numbers were then random ly assigned to either group one or two pre-operatively. Thus prior to the pre-operative interview the researcher knew w hether the patient would be in group 1 (control group) or group 2 (treatm ent group). Each patient was interviewed 24 to 48 hours pre- operatively. Group I At the p re-o p erativ e interview the patients in group 1 were instructed that they should try and cough regularly. In addition to coughing, they should, together with the help o f the nursing staff, get out of bed and walk around the ward on the second post-operative day. They were inform ed that a physiothera­ pist and the medical staff throughout their post-operative hospital stay would routinely assess them. Group 2 The patients in group 2 were also inter­ viewed pre-operatively and they were instructed that they w ould be seen daily after their operation. These patients received their first physiotherapy treatment after extubation. Physiotherapy consisted o f the patient in supine with a m inim um o f 45 degrees head and trunk elevation. The physio­ therapist taught the patient lateral costal expansion exercises follow ed by a pas­ sive expiration and a rib spring at the end o f expiration to stim ulate a deep inspiration at the first post-operative treatm ent session.' The patients were also taught supported coughing using a pillow against the incision site. The com bination o f breathing and coughing exercises were continued until the phy­ siotherapist assessed the chest as cli­ n ically c lear on auscu ltatio n (good breath sounds and no adventitio u s sounds), and an effective unproductive cough was present. The above treatm ent was carried out twice on post-operative days one and two and once on days three and four. From day two the patients were walked around the ward (45 meters) at each treatm ent session. The patients were also allowed to w alk to the bathroom as required. Measurements Pre-operatively, each p atien t’s age, gen­ der, racial group, height (in meters) and w eight (in kilograms) were recorded and the body mass index calculated. P a st m edical and su rg ical history, sm oking history, present history and p re-o p e ra tiv e m ed icatio n w ere also ascertained. Sm oking histories were recorded as pack-years ((num ber o f years sm oked x num ber o f cigarettes per day) / 20) and the length o f sm oking cessation was recorded for those patients who had stopped smoking. Oral tem peratures and respiratory rates were recorded pre-operatively and on days one to four post-operatively. The 06h00 tem perature was recorded from the medical record and the respira­ tory rate was counted over one full m inute prior to the physiotherapy treat­ ment. Earlobe capillary blood (Spiro and D owdeswell, 1976) was taken pre- operatively and on days one and four post-operatively. All measurements were done with the patient in long sitting in bed and breathing room air. I f the patients were on supplem ental oxygen this was discontinued for 15 minutes p rior to earlobe capillary blood gas sampling. For the patients in the group 2 the blood gases were sampled one hour after the morning physiotherapy treat­ ment. Blood gases were analysed using a C iba-C om ing 288 blood gas system machine. The partial pressure of oxygen ’ in arterial blood ( P a 0 2), the partial pres- “ sure o f carbon dioxide in arterial blood (P a C 0 2) and the percentage saturation were recorded. From this the alveolar - arterial oxygen difference (P (A -a )0 2) was calculated. C hest X-rays were taken pre-opera- tively, and on days one, two and four post-operatively. The chest X-rays were scored by a radiologist who was blind to the com position o f the patient groups. The X-ray score was calculated in the ' follow ing way. Each lung was scored individually and then the totals were added together. The location and type o f Abnormality was recorded and the pres­ ence o f pleural effusions was docu­ mented. The scores were allocated as follows: 0 , no abnorm alities were noted and the lung fields were assessed to be radiologically clear, 3, minor collapse / consolidation at one base, involving 1-3 bronchopulm onary segm ents, 7, p ro ­ nounced collapse / consolidation and / or both at one base involving an entire lobe, 15, bilateral collapse and / or con­ solidation and / or patchy infiltrates were noted. ; The researcher noted the length of post-operative stay in hospital and the incidence o f pulm onary com plications (as defined by the criteria below). For the purposes o f this study a pulm onary com plication was defined as: a tempera- ' ture greater than 38.5 degrees Celsius, radiological evidence o f consolidation or collapse and evidence o f respiratory infection clinically and on auscultation as decided by the cardiothoracic sur­ geon. ̂ T-he presence o f all three o f these criteria were required in order to diag­ nose a pulm onary com plication. The cardiothoracic surgeon was unaware as to w hich group the patient was in. Statistics T he data collected was statistically ana­ lysed using the M ann-W hitney sum test, the W ilcoxin test and the chi-square test. M eans, standard deviations and fre­ quency distributions were used to sum ­ marise the data and a p - value equal to or less than 0.05 was considered significant. Intra-examiner reliability Ten percent o f the chest X -rays were scored again by the sam e radiologist six weeks after the initial scoring. The se­ cond set o f scores corresponded 100% ' with the initial set thus proving intra- '' exam iner reliability. RESULTS T he groups were well m atched for age (p = 0 .68), height (p = 0.16), weight 26 S A Jo u r n a l o f Physiotherapy 2000 V o l 56 No 3 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. ) Table 3. O ra l temperatures, a rterial blood gas analyses and chest X -ray scores. G roup 1 G roup 2 PRE-OPERATIVE Temperature (°C) 36 .4 4 "0 .2 1 36 .4 5 " 0.3 5 Respiratory rate (breaths/m in) 16 " 7.14 14.75 " 3.51 Arterial blood gases P a 0 2 78 .7 5 "1 6 .4 3 7 8 .3 2 " 11.90 % Saturation 95 .2 6 "0 .8 3 95 .6 5 " 2.52 P (A -a )0 2 32 .7 9 " 15.82 32.28 " 11.35 Chest X-ray scores Total score 0 0.1 9 " 0 . 7 5 RLL 0 0.1 9 " 0 . 7 5 LLL 0 0.1 9 " 0 . 7 5 DAY 1 POST-OPERATIVE Temperature (°C) 37 .63 " 0.61 37 .9 4 " 0.5 2 Respiratory rate (breaths/min) 23 .93 " 3.81 22.81 " 5.86 Arterial blood gases P a 0 2 71 .3 8 " 17.27 6 0 .3 5 " 14.09 % Saturation 9 8 .1 4 " 0.83 9 5 .6 5 " 3.63 P (A -a )0 2 3 8 .2 0 " 17.62 4 9 .1 3 " 14.09 Chest X-ray scores Total score 1.5 " 2.28 1.69 " 2.18 RLL 1.07 " 1.49 1.13 " 1 . 5 LLL 0.4 3 " 1.09 0 .5 6 " 1.21 DAY 4 POST OPERATIVE Temperature (°C) 3 6 .8 4 " 0.38 36 .8 6 " 0.4 2 Respiratory rate (breaths/min) 22 .8 6 " 4.6 2 21 .8 6 " 6.5 9 Arterial blood gases PaC>2 58.91 " 9.27 6 0 .3 5 " 1 4 . 1 0 % Saturation 92 .0 2 " 4.6 4 91 .98 " 4 . 0 9 P (A -a )0 2 53 .1 3 " 8.83 5 4 .2 4 " 7.8 7 Chest X -ray scores Total score 1.71 " 1.94 2.4 4 " 2.25 RLL 1.5 " 1.56 1.69 " 1.54 LLL 0.21 " 0.8 0 0.7 5 " 1.34 All values are given as means ± standard deviations The number o f patients in treatment group was 16 for all values and in the non-treatment group 14 for all values. RLL, right lower lobe, LLL, left lower lobe, PaC>2 , partial pressure o f oxygen in arterial blood; P(A-a)C>2 , the alveolar - arterial oxygen difference calculated as follows: P (A -a )0 2 = {F I0 2(620 - 47) - P aC 02/ 0 . 8 } - P a 0 2 (p= 0.74), body mass index (p = 0.28) and race (p = 0.23) (see Tables 1 and 2). The P a 0 2 was not significantly different betw een the groups (p = 0.68 pre-opera­ tive, p= 0.07 day 1 and p = 0.92 day 4). This value did change significantly from its pre-operative value to its post-opera­ tive value for both groups. This signifi­ cant decrease in P a 0 2 improved slightly from day one to day four, how ever a profound hypoxia was still present on day four post-operatively in all patients (Table 3). Chest X-ray changes were also not significantly different between the groups (p > 0.05). There was how ever a signi­ ficant difference (p = 0 .001) betw een the pre-operative chest X-rays and the day 1, day 2 and day 4 chest X-rays for both the groups. There was no significant differ­ ence between the day 1 and day 2 X-rays (p = 0.109), the day 1 and day 4 X-rays (p = 0.234) or the day 2 and day 4 chest X-ray scores (p = .0.1). The atelectasis present post-operatively was still pre­ sent on day four regardless o f whether the patient had had physiotherapy or not (Table 3). The incidence o f pulm onary com pli­ cations was 0 % as not a single patient developed a pulm onary com plication as defined by the criteria in this study. The length o f post-operative hospital stay was also not significantly different between the groups. DISCUSSION AND CONCLUSION This study showed that the incidence and severity o f hypoxaem ia, chest X-ray changes, pulm onary com plications and length o f post-operative hospital stay were not significantly different for a control group who received a pre-opera­ tive instruction and no prophylactic chest physiotherapy in patients under­ going uncom plicated cardiac valve sur­ gery. The deleterious changes observed in arterial blood gases, chest X-rays and tem perature were not o f clinical signi­ ficance as no patients were diagnosed as having a post-operative pulmonary com plication. ' The results of this study although conducted on a sm aller patient popula­ tion seem to confirm the results o f the Stiller et al (1994 and 1995) and Jenkins et al (1989 and 1990) studies. It would ' thus seem that chest physiotherapy has no role to play in the prevention of p ul­ m onary com plications after uncom pli­ cated cardiac valvular su rg ery .' These results do not attem pt in any way to draw conclusions for patients undergo­ ing em ergency valvular surgery or for patients with pre-existing pulm onary disease. For both these patient groups further studies will need to be conducted. In addition this study makes no attem pt to define the role o f p h y siotherapy in patients with prolonged intubation times or intensive care stays, or in those who develop post-operative neurolo­ gical complications. It is possible that the patients in group 1, by virtue o f signing informed consent and talking to patients who were being treated may have altered their post-oper­ ative behaviour. This may have acted as a confounding variable. It would seem then that prophylactic chest physiotherapy in the post-opera­ tive valvular surgery patient is not indi­ cated provided the patient is mobilised SA Jo u r n a l o f Physiotherapy 2000 V o l 56 No 3 27 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. ) out o f bed as soon as possible and instructions to cough post-operatively are given. It is possible that this change in p h y sio th erap y p ractice has been bou g h t about due to m ore efficient anaesthetic techniques, intra-operative and post-operative patient care. It is the recom m endation o f this study that phy­ siotherapists assess all patients post- operatively. Treatm ent is only indicated if and when a pulm onary com plication is identified. REFERENCES Barlow JB 1992(a) Aspects o f active rheuma­ tic carditis. Australian and N ew Zealand Journal 22: 592 - 600. Barlow JB 1992(b) Idiopathic (Degenerative) and R heum atic Mitral V alve Prolapse: Historical A spects and an O verview . The Journal o f Heart Valve D isease '!': 163 - 174 Jenkins SC, Soutar SA , Loukota JM, Johnson LC, M oxham J 1989 Physiotherapy after coro­ nary artery surgery: are breathing exercises necessary? Thorax 44: 6 34 - 639 Jenkins SC, Soutar SA , Loukota JM, Johnson LC, M oxham J 1990 A comparison o f breath­ ing exercises, incentive spirometry and m obi­ lisation after coronary artery surgery. Physiotherapy Theory and Practice 6: 117 - 126 Johnson D, Thompson D , M ayers 1 1996 The effect o f physical therapy on respiratory com plications follow in g cardiac valve surgery. Chest 109: 638 - 6 4 4 Marcus RH, Sareli P, P ocock WA, Barlow JB 1994 The spectrum o f severe rheumatic mitral valve disease in a developing country. Correlations among clinical presentation, sur­ gical pathologic findings, and haem odynam ic sequelae. Annals o f Internal M edicine 120: 177 - 183. M cLaren MJ, H aw kin s D M , K oornhof H J,Bloom KR, Bramwell-Jones DM , Cohen E, Gale GE, Kamarek K, Lachman A S, Lakier JB, P ocock WA, Barlow JB 1975 Epidem iology o f rheumatic heart disease in black school children o f S ow eto, Johannesburg. British M edical Journal 3: 47 4 - 478. Spiro SG and D o w d e sw e ll IRG. 1976 Arterialised ear lobe blood samples for blood gas tensions. British Journal o f D iseases o f the Chest 70: 263 - 268 Stiller K, M ontarello J, W allace M, D aff M, Grant R, Jenkins S, Hall B, Yates H 1994 Are breathing and coughing exercises n eces­ sary after coronary artery surgery? Physio­ therapy Theory and Practice 10: 143 - 152 S tiller K, Crawford R, M cln n es M, M ontarello J, Hall B 1995 The incidence o f pulm onary com p lic a tio n s in patients not receiving prophylactic chest physiotherapy after cardiac surgery. Physiotherapy Theory and Practice 11: 205 - 208 THE UK’ S LEADI NG A G E N C Y FOR P H Y S I O T H E R A P I S T S WORKING IN THE UK NEEDN’T BE A PUZZLE ♦ Friendly ad vice on: • State R egistration • UK entry • A ccom m od ation ♦ W idest ch o ice o f p osts ♦ Top r a te s o f pay ♦ CPSM fe e s reim b u rsed ♦ FREE Indem nity In su ran ce For our FREE ‘W orking H olidays in B ritain’ b roch u re, co n ta ct Debi Faulder m c s p s r p m e c i on TOLL-FREE 0 8 0 0 - 9 9 - 3 0 5 5 (2 4 Hours) or F ax on 0 9 4 4 181 2 0 7 6 8 9 4 or E-m ail: locu m s@ corin th .co.u k *Subjcct to th e Term s o f th e Policy You’ll be amazed how CORINTH MEDICAL can help! Corinth Medical 5 Theobald Court, Theobald Street Borehamwood, Herts, WD6 4RN, UK 28 SA Jo u r n a l o f -Physiotherapy 2000 V o l 56 No 3 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. ) mailto:locums@corinth.co.uk