The PDF of the article you requested follows this cover page.  

This is an enhanced PDF from The Journal of Bone and Joint Surgery

64:242-248, 1982. J. Bone Joint Surg. Am.K Kumano and N Tsuyama    

Pulmonary function before and after surgical correction of scoliosis

This information is current as of October 15, 2006

Reprints and Permissions

Permissions] link. and click on the [Reprints andjbjs.orgarticle, or locate the article citation on

to use material from thisorder reprints or request permissionClick here to

Publisher Information

www.jbjs.org20 Pickering Street, Needham, MA 02492-3157The Journal of Bone and Joint Surgery

on October 15, 2006 www.ejbjs.orgDownloaded from

242 J. W. PRITCHETT

9-13, Jan. 1976.2. McMANUS,FRANK;RANG,MERCER;and HESLIN,D. J.: Acute Dislocation of the Patella in Children. The Natural History. Clin. Orthop., 139:

88-91, 1979.3. MoRETZ,ALFRED,111;RASKIN,ALAN;and GRANA,W. A.: Oklahoma High School Football Injury Study: A Preliminary Report J Oklahoma

State Med. Assn., 71: 85-88, 1978.4. PRITCHETT,J. W.: High Cost of High School Football Injuries. Am. J. Sports Med., 8: 197-199, 1980.5. SONNE-HOLM,S.; AHN, N. C.; and FLEDELIUS,I.: Meniscectomy, A 10-Year Follow-up of 150 Athletes [abstract]. Acta Orthop. Scandinavica,

48: 223, 1977.

Copyright 1982 by The Journal of Bone and Joint Surgery, Incorporated

Pulmonary Function before andafter Surgical Correction of Scoliosis*

BY K1YOSH1 RUMANO, M.D.t, AND NAOICHI TSUYAMA, M.D.t, TOKYO, JAPAN

From the Tokyo University Hospital and the Toranomon Hospital, Tok\o

ABSTRACT:We compared the preoperative andpostoperative pulmonary function of thirty-one scolioticpatients ranging in age from nine to twenty-five years.The mean postoperative follow-up period was three

years and eight months. Twenty patients were treated bya posterior procedure with Harrington instrumentation.At more than two years postoperatively a significant improvement in the pulmonary function was noted, particularly in patients with a preoperative curve of lessthan 90 degrees (Cobb angle) and in those in whom thecorrection was greater than 30 per cent. The remainingeleven patients were treated by an anterior procedure,primarily a Dwyer operation, with or without posteriorHarrington instrumentation. These patients manifestedno remarkable improvement in pulmonary functionmore than two years after surgery; three patientsshowed deterioration. The results of tests performed lessthan two years postoperatively showed no improvementin pulmonary function, irrespective of the types of assessments used.

We attribute our long-term improvements to ashortening of the postoperative period of plaster-castimmobilization and to the use of a plastic corset whichallowed relatively free chest motion. We suggest that theDwyer operation should be restricted to patients with asevere spinal deformity.

Previously reported observations regarding the car-

diopulmonary function and the natural history of scolioticpatients1"3'5'6'18'19'22"24clearly indicate that severe scoliosis

greatly affects cardiopulmonary function and that, particularly in the surgical treatment of scoliosis, the primary aimshould be to improve this function.

In the present study we evaluated the effect of surgicalcorrection of scoliosis on the pulmonary function of thirty-

* Read in part at the Annual Meeting of the Japanese OrthopaedicAssociation, April 11, 1979.

t Department of Orthopaedic Surgery, University of Tokyo, Hongo7-3-1, Bunkyo-ku, Tokyo 113, Japan.

one patients by comparing the results of preoperative andpostoperative routine pulmonary-function tests and arterialblood-gas analyses.

Material

From 1973 to 1978, forty-six scoliotic patients under

went surgery at the Tokyo University Hospital and at theToranomon Hospital. Of these, fifteen were excluded fromthe present study because they were younger than six yearsold at the time of surgery or because of insufficient availabledata. The remaining thirty-one (eighteen female and thirteen

male) patients were last examined not less than two yearspostoperatively. There were twenty patients with idiopathicscoliosis, four who had had poliomyelitis, one paraplegicpatient, two with congenital scoliosis, two patients withneurofibromatosis, one with a myopathy, and one with Mar-fan's syndrome (Tables 1 and II).

The mean age at the first preoperative examination wasseventeen years and four months (range, nine to twenty-five

years). At the last postoperative examination the mean agewas 20.5 years. The mean postoperative follow-up period

was three years and eight months.

Operative MethodsThe thirty-one patients were divided into two groups

according to the operative procedure employed. There weretwenty patients who had a posterior procedure. Harrington-

rod instrumentation and posterior spine fusion with iliacbone-grafting17 was the operation primarily used in this

group. Of these twenty patients, four underwent additionalsubsequent surgical procedures: one had a laminectomy andthree, a rod reinsertion (Table I). There were eleven patientswho had an anterior procedure; the types of operations usedare shown in Table II.

Most patients were treated preoperatively with someform of traction; however, no specific breathing exerciseswere instituted preoperatively or postoperatively. A Risserplaster cast was applied two weeks after the Harrington instrumentation and the patients then were allowed to walk.

THE JOURNAL OF BONE AND JOINT SURGERY on October 15, 2006 www.ejbjs.orgDownloaded from

PULMONARY FUNCTION BEFORE AND AFTER SURGICAL CORRECTION OF SCOLIOSIS 243

TABLE 1

PATIENTSWITHAPOSTERIORPROCEDURE

Case1234567891011121314151617181920MeanSexFMFMFFFMFFMFFFFFFFFFDiagnosisCongen.

scoliosisIdiopath.scoliosisIdiopath.

scoliosisIdiopath.

scoliosisIdiopath.scoliosisIdiopath.scoliosisIdiopath.scoliosisIdiopath.scoliosisIdiopath.scoliosisIdiopath.scoliosisNeurofibromat.Idiopath.

scoliosisIdiopath.scoliosisIdiopath.scoliosisIdiopath.scoliosisMyopathyIdiopath.

scoliosisIdiopath.

scoliosisIdiopath.scoliosisMarian's

syndromeCurve

PatternT7-L2T7-L2T7-L1T5-T10T4-T11T5-L1T6-T12T8-L2T11-L3T4-T8/T8-L4T7-T10T5-T12T5-T11T4-T11/T11-L4T5-T11T9-L3T11-L3T7-T12T6-T12T6-T11/T11-L4Cobb

Angle(Degrees)

Preop.Postop.10610565939195571085947/5827617262/491165248876452/307064674562557034764942/4215333340/321003735663544/2848Correction(Percent)402731334027403017224546543514292725451232Age

at InitialExamination

(Yrs. +Mos.)15

+614+1114

+819

+219+614+417+012+1118+817+1115+1215+214+415+112+615

+220+214

+815+214+815+ 5Length

ofFollow-up

(Yrs. +Mas.,6

+45+115+105

+65+54+104

+94+33+63+33+12+112+102

+92+42

+42+32

+32+12+13+ 8Procedure1Harrington

instrument.Laminect.,Harringtoninstrument.Harrington

instrument.;rod reinsert. , bonegraftHarrington

instrument.Harringtoninstrument.Harringtoninstrument.Harringtoninstrument.Harringtoninstrument.Harringtoninstrument.Harringtoninstrument.Harringtoninstrument.Harringtoninstrument.Harringtoninstrument.Harringtoninstrument.Harrington

instrument.;rod reinsert. , bonegraftHarrington

instrument.Harringtoninstrument.;

rod reinsert. , bonegraftHarringtoninstrument.Harringtoninstrument.Harrington

instrument.

The cast was worn for three to four months, and then wasreplaced by a plastic corset which was worn for the remainder of the first postoperative year. During this time the patients were allowed to take showers two or three times aweek and deep breathing was encouraged.

A Dwyer procedure and anterior fusion was done via atranspleural and retroperitoneal exposure7. In patients un

dergoing combined anterior and posterior procedures, theanterior procedure was usually performed first, followedseveral weeks later by Harrington instrumentation. Postoperative management was as described for the patients whohad undergone the Harrington operation.

All patients discarded their corsets during the secondpostoperative year and no sports activity was permitted untilafter the second postoperative year.

Harrington instrumentation was the only surgical procedure that was performed primarily in patients withidiopathic or other types of scoliosis in which the curve wasflexible and measured less than 90 degrees (Cobb angle).Combined anterior and posterior procedures were carriedout in patients with severe paralytic or idiopathic scoliosismanifesting a severely rotated and rigid curve.

Evaluation of Pulmonary Function

Preoperative pulmonary evaluation was performedwithin a few weeks before surgery, and postoperative evaluation was started one year after surgery and repeated an

nually for as long as possible. The lung volume was measured on a Godait spirometer, using the helium dilutiontechnique, with the patient in the sitting position4. Pulmo

nary ventilation was measured with the patient standing upright, and a 13.5-liter Benedict-Roth spirometer was em

ployed . To compare preoperative and postoperative values,the percentage of predicted normal values15'16 was used.

According to the suggestions of Gazioglu et al. and Hepperet al., the corrected rather than the actual height of the patient was utilized in the calculations, and the arm span wasused for the estimation. The formula to obtain the correlation between arm span and body height was derived fromour survey of 400 children between the ages of six and seventeen years.

boys:height = 0.946 x arm span + 6.308 (standard deviation = 3.29)

girls:height = 0.926 x arm span + 9.856 (standard deviation = 3.45)

Arterial blood samples, obtained from the patients afterthey had remained in a recumbent position for ten minutes,were analyzed on an Astrup analyzer.

Results

For the twenty patients treated by the posterior procedure, the mean age was fifteen years and five months at the

VOL. 64-A, NO. 2, FEBRUARY 1982 on October 15, 2006 www.ejbjs.orgDownloaded from

244 KIYOSHI KUMANO AND NAOICHI TSUYAMA

TABLE II

PATIENTSWITHANANTERIORPROCEDURE

Case2122232425262728293031MeanSexMMFMMMMMFMMDiagnosisPost-polio,

scoliosisIdiopath.

scoliosisIdiopath.

scoliosisCongen.scoliosisIdiopath.

scoliosisPost-polio,scoliosisNeurofibromat.Post-polio,

scoliosisIdiopath.

scoliosisParaplegiaPost-polio,

scoliosisCurve

PatternT7-L2T6-L2T11-L4T10-L4T5-L2T4-T11/T11-L5T6-L2T9-L4T6-L1T10-L5T6-T12Cobb

Angle(Degrees)

Preop. Postop. Correction(Percent)8090645511767/9070124150105/14114110054648338637/4837828589/869262293183402642474445294442Age

at InitialExamination

(Yrs. +Mos.)15

+25

+13

+14+22

+9+13

+17

+23

+16

+20+17

+101165810771426Length

ofFollow-up

(Yrs. +Mos.6

+34

+84

+54+23

+63+52

+92

+92

+52

+42+03

+ 9Procedure;Post,

osteot.,HarringtonAnt.

fusion*,instrument.Dwyer

proc.ant.

fusion*,instrument.HarringtonHarrington

instrument.,ant.fusion*Dwyer

proc.Dwyerproc.;instrument.bone

graftDwyerproc.,instrument.Dwyer

proc.,instrument.Post,

osteot..HarringtonDwyer

proc.Dwyerproc.,instrument.Harrington;

rodreinsert.,HarringtonHarringtonDwyer

proc.,instrument.Harrington

* Without instrumentation.

time of the initial examination. The mean follow-up period

in this group was three years and eight months. The meanCobb angle was 70 degrees preoperatively and 48 degreespostoperatively. The mean correction was 32 per cent(Table I).

For the eleven patients treated by the anterior procedure, the mean age at the initial examination was 17.5years, and the mean follow-up period was three years and

nine months. The mean Cobb angle was 100 degreespreoperatively and 63 degrees postoperatively. The meancorrection was 42 per cent (Table II).

In all but two patients in each group (Cases 15, 20, 26,and 27) skeletal growth had been completed at the last postoperative examination.

The preoperative and postoperative results of therespiratory function tests are summarized in Table III.

To test whether individual patients had improvementpostoperatively, a scoring system was devised. Each respiratory parameter showing postoperative improvement of atleast 15 per cent was assigned a value of +1. A value of-1 was assigned to parameters showing a 15 per cent

postoperative decline. If the postoperative arterial oxygenpartial pressure was at least ten millimeters of mercuryabove the preoperative level, a +1 value was given. A —1

value was assigned when the postoperative level was atleast ten millimeters of mercury below the preoperativelevel. Otherwise, respiratory function or arterial oxygenpartial pressure was scored as zero. The total score forpulmonary function was then determined: if it was positive, it indicated improvement; if it was negative, it revealed a postoperative worsening of the patient's condi

tion; and if it was zero, no postoperative change had occurred.

At less than two years postoperatively, there were nosignificant changes from the preoperative values in the patients who had a posterior procedure. However, the vitalcapacity in those with an anterior procedure was significantly (p < 0.05) lower than the preoperative values(Table III).

The same studies were repeated more than two yearspostoperatively. As shown in Table III, patients who hadbeen treated with a posterior procedure showed a significant increase in vital capacity, total lung volume,functional residual capacity, maximum ventilatory volume, and maximum mid-flow rate. On the other hand, no

such findings were obtained in those who had had an anterior procedure. There was no significant difference between the preoperative and postoperative levels of arterialoxygen partial pressure in either group of patients. Elevenof the twenty patients with a posterior procedure showedmore than a 30 per cent correction of the curve. Table IVshows the results of the respiratory function tests thatwere performed more than two years postoperatively inpatients classified according to the amount of curve correction. Better results were obviously obtained in patientswith a higher percentage (more than 30 per cent) of correction.

The twenty patients with a posterior procedure weresubdivided into two groups according to whether thepreoperative Cobb angle was more or less than 90 degrees.As shown in Table IV, the thirteen patients with an angleof less than 90 degrees had a statistically significant im-

THE JOURNAL OF BONE AND JOINT SURGERY on October 15, 2006 www.ejbjs.orgDownloaded from

PULMONARY FUNCTION BEFORE AND AFTER SURGICAL CORRECTION OF SCOLIOSIS 245

TABLE III

RESPIRATORYFUNCTIONANDARTERIALOXYGENPARTIALPRESSURE

TotalLungPosterior

procedurePreop.Postop.,

< 2yrs.Postop.,> 2yrs.Anterior

procedurePreop.Postop.,

< 2yrs.Postop.,> 2 yrs.No.

ofPatients20142011811VitalLiters2.32.32.62.72.32.7CapacityPerCent646571*6552t60ResiduiLiters0.91.01.11.21.51.3il

VolumePerCent858890126129103CapacityLiters3.23.33.73.83.73.9PerCent697076f756771FunctionalResidual

CapacityLiters1.71.82.02.02.32.1PerCent707380t808379Maximum

Voluntar)1VentilationLiters687787808692.7Per

Cent546270*665860Maximum

ExpiratoryMid-FlowRateLiters2.52.82.92.62.92.7Per

Cent596471*506057ArterialOxygenPartial

Pressure(mmHg)899590909089

* Highly significantly different from preoperative value (p < 0.01).t Significantly different from preoperative value (p < 0.05).

provement in respiratory reserve, while the other sevenpatients did not.

However, based on our scoring method, seventeen ofthe twenty patients with the posterior procedure showedsome improvement in postoperative pulmonary function.The remaining three patients showed no change. Of theeleven patients treated by the anterior procedure, fourshowed some (not statistically significant) improvement,four revealed no change, and three had worsening.

Over-all, of the thirty-one patients in our series,twenty-one (67.7 per cent) showed postoperative im

provement, seven (22.6 per cent) demonstrated no change,and three (9.7 per cent) worsened postoperatively. Two ofthe three patients whose condition worsened postoperatively (Cases 23 and 25) had been treated by a Dwyer operation for idiopathic scoliosis and one (Case 27), a patientwith neurofibromatosis, had been treated by a combinationof a Dwyer procedure and Harrington instrumentation.

Discussion

There have been several studies of the effect on pulmonary function of scoliosis and its treatment. In theirclassic report, Bergofsky et al. studied pulmonary functionin twenty-seven patients, mostly forty years old or more,

with severe kyphoscoliosis; they concluded that the frequent occurrence of cardiorespiratory failure in this disease was due to alveolar hypoventilation and cor pul-

monale. The hypoventilation was characteristically attributable to the thoracic deformity of the kyphoscoliosis.The cor pulmonale was primarily the result of pulmonaryhypertension due to vessels being compressed in thecharacteristically poorly developed, smaller-than-normal

lungs of these patients. Ting and Lyons reported that thecompliance of the respiratory system in scoliotic patientswas approximately one-third of normal, mostly because of

a decrease in thoracic compliance. Both studies suggestedthat late cardiopulmonary failure in scoliotic patients isprobably attributable to a decrease in compliance of thethorax.

On the other hand, Caro and DuBois studied sevenscoliotic children, seven to fifteen years old, who hadnormal thoracic compliance but decreased compliance ofthe lungs. They suggested that the rigidity of the chestcage was a late complication of scoliosis. Early in thecourse of the disease there was decreased lung compliance, probably due to obstruction of the peripheral airway. The authors reproduced the pulmonary changes theyfound in scoliosis by tightly strapping the chests of normal

TABLE IV

CORRELATIONOF RESPIRATORYFUNCTION, AMOUNTOF CURVE CORRECTION,ANDCOBB ANGLE*!

Curvecorrection>30

percent<30 percentPreop.

Cobbangle>90degrees(mean,

102degrees)<90degrees(mean,

56 degrees)No.

ofPatients119713VitalCapacity8t548tResidualVolume14-805Total

LungCapacitylit167ÕFunctional

ResidualCapacity18t3513*Maximum

VoluntaryVentilation23*41119*Maximum

ExpiratoryMid-FlowRate19t21510*

* The patients with the anterior procedures were excluded.

t The values in the table represent the postoperative increase; that is, the increase in the percentage of the predicted normal value after operation.For the method of assessment, see text,

t P < 0.01.

VOL. 64-A, NO. 2, FEBRUARY 1982 on October 15, 2006 www.ejbjs.orgDownloaded from

246 KIYOSHI RUMANO AND NAOICHI TSUYAMA

men and causing parts of the lung to become occluded toventilation. They emphasized that the aim of orthopaedictreatment of kyphoscoliosis should be to correct the spinalcurve without further interfering with the expansion of thechest cage. Any long-term restriction of chest expansion,as in a tight plaster body jacket, might lead to permanentocclusion of additional regions of the lung from ventilation, or to a decrease in the mobility of the chest wall.Bjure et al. proved that closure of the peripheral airwaysoccurs in the lungs of scoliotic patients as shown by 133xe-non bolus-injection techniques and as Caro and DuBoissuggested. In the series of Bjure et al., in six of nineteensubjects who were younger than forty years old, closure ofthe airways began at lung volumes greater than thefunctional residual volume, the usual resting expiratorylevel, whereas it normally occurred at a lung volumebelow the functional residual volume in patients of thatage. The alveoli that were closed, and therefore unable totake in oxygen, created a shunting effect, air being shuntedto those alveoli that were still open.

Dollery et al. and Shannon et al., using intravenousbolus injections of the non-soluble isotope 133xenon in

scoliotic patients, noted hypoventilation and decreasedperfusion restricted to the basal lung fields compared withthe upper lung fields. In normal subjects there was progressively increasing ventilation craniocaudally. These researchers attributed their findings to the compression ofthe extra-alveolar vessels in the deformed thorax of thescoliotic patients. They suggested that the vascular elasticity decreases and the degree of compression increases asthe severity of the deformity becomes more pronounced,eventually leading to irreversible changes in the pulmonary vessels.

Shannon et al. showed that following surgical correction of kyphoscoliosis the physiological pulmonary deadspace was decreased by 40 per cent, but in patients with aCobb angle of greater than 65 degrees there was no significant postoperative improvement in the regional perfusion of the lungs. They attributed this finding to a possiblyirreversible pulmonary change and suggested that surgeryshould be performed before the angle of curvature reaches70 degrees.

Based on these previous studies, it is apparent that toimprove the pulmonary function of scoliotic patients onemust relieve the compression on the pulmonary parenchyma due to the deformation of the thorax before irreversible changes in the vessels occur. Surgical stabilization ofthe spine and correction of the deformed thorax shouldserve this purpose.

At present there are two general points of view regarding the effect of spine fusion in patients with scoliosis.The opinion most widely held is that it averts the progressive deterioration of pulmonary function which, accordingto previous reports, worsens as the severity of an abnormalcurve increases13>17>21.Some authors have believed that

surgery actually improves the pulmonary function.Gucker, Takahashi, and Makley et al. subscribed to the

former opinion, but it must be recognized that their resultswere all obtained before Harrington instrumentation wasintroduced. As to the latter idea, Westigate and Moestudied seventy-four patients between the ages of four andforty years old with diagnoses of post-poliomyelitic,idiopathic, and congenital scoliosis. The mean Cobb anglein their series was 87 degrees. All of their patients weretreated by Harrington instrumentation. In thirty-one ofthem a comparison of the preoperative and postoperative(more than one year after surgery) vital capacity andmaximum breathing capacity showed that there was apostoperative decrease in both in most patients. In ten oftheir patients the arterial oxygen saturation improved froma preoperative value of 93.5 per cent to a postoperativevalue of 95.2 per cent. Therefore, Westigate and Moecame to the conclusion that the Harrington method did notimprove the average vital capacity or the maximum breathing capacity, but could improve somewhat the arterialoxygen saturation.

Meister and Heine noted no improvement in the vitalcapacity, total lung volume, residual volume, or arterialoxygen partial pressure one year after Harrington instrumentation in thirty idiopathic scoliotic patients whowere between the ages of twelve and nineteen years andhad a mean preoperative Cobb angle of 79 degrees. Theystressed the importance of surgical intervention before deterioration of pulmonary function occurs.

Zorab et al. reported the absence of any real improvement in the lung volume of forty-three scoliotic patients (eleven to twenty years old) with a mean Cobb angleof 92 degrees who were followed for a prolonged period.However, they gave no details on the type of surgery performed on their patients.

Gazioglu et al., on the other hand, reported improvement in pulmonary function in thirty-three patients withidiopathic scoliosis whose mean age was fifteen years oldwhen they were operated on. They were all treated byHarrington instrumentation and had a mean preoperativeCobb angle of 62 degrees. The postoperative angle averaged 21 degrees, so that mean correction was 67 per cent.Seventeen of these patients manifested abnormal pulmonary function preoperatively, and postoperatively twelveof them showed improvement. The remaining sixteen patients had normal pulmonary function both preoperativelyand postoperatively. The vital capacity and total lungcapacity of these patients increased postoperatively bymore than 18 per cent. The maximum expiratory mid-flowrate increased by 15 per cent. Gazioglu et al. stated that thecorrection of the spinal curve was the factor that improvedthe pulmonary status in their patients.

Lindh and Bjure studied eighty-seven patients withidiopathic scoliosis and five with paralytic scoliosis whosemean age was fifteen years at the time of spine fusion. Themean preoperative curve was 72 degrees in the patientswith idiopathic scoliosis and 99 degrees in those withparalytic scoliosis. The average correction was 46 percent. There was a 10 per cent average increase in lung vol-

THE JOURNAL OF BONE AND JOINT SURGERY on October 15, 2006 www.ejbjs.orgDownloaded from

PULMONARY FUNCTION BEFORE AND AFTER SURGICAL CORRECTION OF SCOLIOS1S 247

urne more than eighteen months postoperatively in sixty-three patients, all of whom were treated by Harrington instrumentation.

In our series, the twenty patients who had Harringtonprocedures were nine to twenty-three years old when operated on, with a mean age of fifteen years. All of themwere observed more than two years postoperatively. Themean preoperative Cobb angle was 70 degrees and thepostoperative angle was 48 degrees. While we realize thatour patients had slightly more severe curves and more impaired pulmonary function compared with the patients ofGazioglu et al. and of Lindh and Bjure, we believe that ourseries was essentially similar. Our series, like theirs,showed statistically significant improvement in the pulmonary function tests more than two years after the surgerywas done.

With reference to the changes in pulmonary functionover time, both Westigate and Moe and Gazioglu et al.noted no significant difference in postoperative pulmonaryfunction of the scoliotic patients when comparing findingsat one year with those at more than one year after surgery. Lindh and Bjure reported that after a prolongedfollow-up, pulmonary function in patients treated by Harrington instrumentation showed a tendency to improvement. Our study showed that when the same patients weretested less than and more than two years postoperatively,better results were obtained at the last examination, irrespective of whether the fusion was posterior or anterior(Table IV). These findings support our clinical impressionthat, in most patients, at least two years is required beforeoptimum improvement is reached. The unsatisfactory results that Meister and Heine obtained might not have persisted if their patients had been followed for more than twoyears postoperatively.

Our patients wore a Risser cast for three to fourmonths postoperatively, so the duration of postoperativeplaster-cast immobilization was shorter than that reportedin other series17. Our use of a special plastic body corset

for the remainder of the first postoperative year was designed to facilitate relatively free movement of the chest.

Lindh and Bjure advocated the use of a Milwaukeebrace postoperatively and attributed the improvement inpulmonary function in their patients largely to the use ofthis device. This type of postoperative management mayindeed have beneficial effects on pulmonary function.Caro and DuBois indicated that a tight plaster cast mightpermanently occlude more regions of the lungs to ventilation than were occluded because of the deformity itself.

In our series of patients who had Harrington instrumentation, better results were obtained in those inwhom there was a higher percentage of correction of theCobb angle. Gazioglu et al., however, found no correlation between the amount of correction and the postoperative improvement in pulmonary function. This differencein findings may be explained by the differences in the patients in our series compared with theirs. They had morepatients with normal preoperative pulmonary function andrelatively less severe curves.

However, in contrast, Lindh and Bjure reported betterresults in patients with a preoperative Cobb angle of morethan 90 degrees than in those in whom this angle was lessthan 90 degrees, a result that is opposite to our findings.We cannot explain this discrepancy in results. Althoughthe number of our patients is too small to view our resultsas conclusive, our findings are in accord with those ofShannon et al. who suggested that in patients with a curveof more than 70 degrees permanent damage to the pulmonary circulation may have supervened.

In our group of patients who had anterior proceduresthere were more patients with a severe spinal deformityand paralytic scoliosis than in the group with posteriorprocedures and, in addition, a mean of two operationswere performed in the anterior-procedure group. Thesetwo groups of patients, therefore, cannot be compared asto the effectiveness of the two types of operation in improving pulmonary function.

Nevertheless, it should be emphasized that there wasa remarkable difference in results between the two groups:no postoperative deterioration of pulmonary function wasnoted in the patients with the posterior procedure, while ofeleven patients treated by the anterior procedure, three(Cases 23, 25, and 27) who had undergone a Dwyer procedure showed distinct worsening of pulmonary function.In Cases 23 and 27 this may be explained by the fact thatthe kyphosis was exacerbated postoperatively. In Case 25our explanation is that atelectasis occurred intraopera-tively, necessitating tracheostomy. These two complications may have been responsible for the deterioration inpulmonary function observed postoperatively in thesethree patients; the complications were not observed in theremaining patients with the anterior procedure.

There have been, to our knowledge, no previouslypublished studies on pulmonary function after anteriorspine fusion (the Dwyer procedure). Considering the extent of surgical intervention involved, including transversesection of muscles (the latissimus dorsi muscle and the diaphragm in particular) and resection of at least one rib, andconsidering the likelihood of postoperative pleural adhesions and atelectasis in some patients7, we suggest that

Harrington instrumentation should be preferred as the procedure of choice in most patients.

Admittedly, the Dwyer procedure is most efficaciousin stabilizing the collapsing spine in patients with more severe curves such as are seen in post-poliomyeliticscoliosis20. In three of our patients (Cases 28, 30, and 31)

the curve exceeded 120 degrees, and they were treated bythe Dwyer procedure. Their postoperative tests showed nodeterioration of pulmonary function and, in fact, one ofthem (Case 28) demonstrated an improvement in the regional lung perfusion, as evidenced by technetium-99mmacroaggregated albumin scintigraphy, which was performed preoperatively as well as three years after surgery.The number of our patients with the anterior proceduretherefore may not be large enough to allow one to prognosticate on the effects of that procedure on pulmonaryfunction.

on October 15, 2006 www.ejbjs.orgDownloaded from

248 KIYOSHI KUMANO AND NAOICHI TSUYAMA

References1. BERGOFSKY,E. H.; TURINO, G. M.; and FISHMAN, A. P.: Cardiorespiratory Failure in Kyphoscoliosis. Medicine, 38: 263-317, 1959.2. BJURE, J.;GR1MBY, G.; KASALICKU,J.; LINDH, M.; and NACHEMSON,A.: Respiratory Impairment and Airway Closure in Patients with Un

treated Idiopathic Scoliosis. Thorax, 25: 451-456, 1970.3. CARO, C. G., and DuBois, A. B.: Pulmonary Function in Kyphoscoliosis. Thorax, 16: 282-290, 1961.4. COMROE, J. H.; FORSTER, R. E.; DuBois, A. B.; BRISCOE, W. A.; and CARLSEN, ELIZABETH: The Lung. Clinical Physiology and Pulmonary

Function Tests. Ed. 2. Chicago, Year Book Medical, 1962.5. DAVIES, G., and REID, LYNNE: Effect of Scoliosis on Growth of Alveoli and Pulmonary Arteries and on Right Ventricle. Arch. Dis. Child.. 46:

623-632, 1971.6. DOLLERY, C. T.; GiLLiAM, P. M. S.; HUGH-JONES, P.; and ZORAB, P.A.: Regional Lung Function in Kyphoscoliosis. Thorax, 20: 175-181,

1965.7. DWYER, A. F., and SCHAFER, M. F.: Anterior Approach to Scoliosis. Results of Treatment in Fifty-one Cases. J. Bone and Joint Surg., 56-B(2>:

218-224, 1974.8. GAZIOCLU, KADDUSI;GOLDSTEIN, L. A.; FEMI-PEARSE, DEJI; and Yu, P. N.: Pulmonary Function in Idiopathic Scoliosis. Comparative Evalua

tion Before and After Orthopaedic Correction. J. Bone and Joint Surg., SO-A: 1391-1399, Oct. 1968.9. GUCKER, THOMAS:Changes in Vital Capacity in Scoliosis. Preliminary Report on Effects of Treatment. J. Bone and Joint Surg., 44-A: 469-481,

April 1962.10. HEPPER, N. G. G.; BLACK, L. F.; and FOWLER, W. S.: Relationships of Lung Volume to Height and Arm Span in Normal Subjects and in

Patients with Spinal Deformity. Am. Rev. Respir. Dis., 91: 356-362, 1965.11. LINDH, MARGARETA,and BJURE, JAN: Lung Volumes in Scoliosis Before and After Correction by the Harrington Instrumentation. Acta Orthop.

Scandinavica, 46: 934-948, 1975.12. MAKLEY, J. T.; HERNDON.C. H.; INKLEY, SCOTT; DOERSHUK,CARL; MATTHEWS, L. W.; POST, R. H.; andLiTTELL, A. S.: Pulmonary Func

tion in Paralytic and Non-Paralytic Scoliosis Before and After Treatment. A Study of Sixty-three Cases. J. Bone and Joint Surg., 50-A: 1379-1390, Oct. 1968.

13. MANKIN, H.; GRAHAM, J. J.; and SCHACK, JEROME: Cardiopulmonary Function in Mild and Moderate Idiopathic Scoliosis. J. Bone and JointSurg., 46-A: 53-62, Jan. 1964.

14. MEISTER, R., and HEINE, J.: Vergleichen de Untersuchungen der Lungenfunktion bei jugendlichen Skoliose- patienten vor und der Operationnach Harrington. Zeitschr. Orthop. Grenz., 111: 749-755, 1973.

15. MIYAZAWA, M.; YANAI.C.; OKADA, K.; and FUKUDA, S.: Normal Value of Lung Volumes, Pulmonary Ventilation and Diffusion in the Adult.Rinshobyori, 19:415-422, 1971.

16. MIYAZAWA, M.; YANAI, C.; OKADA, K.; and FUKUDA, S.: Lung Volumes and Pulmonary Ventilation in Children. Rinshobyori (Supplement),20: 266, 1972.

17. MOE, J. H.; WINTER, R. B.; BRADFORD,D. S.; and LONSTEIN, J. E.: Scoliosis and Other Spinal Deformities, pp. 467-526. Philadelphia, W. B.Saunders, 1978.

18. NACHEMSON,ALF: A Long Term Follow-Up Study of Non-Treated Scoliosis. Acta Orthop. Scandinavica, 39: 466-476, 1968.19. NILSONNE, ULF, and LUNDGREN, K. D.: Long-Term Prognosis in Idiopathic Scoliosis. Acta Orthop. Scandinavica. 39: 456-465, 1968.20. O'BRIEN, J. P., and YAU, A.C.M.C.: Anterior and Posterior Correction and Fusion for Paralytic Scoliosis. Clin. Orthop., 86: 151-153, 1972.21. RoTHMAN, R. H., and SIMEONE, F. A.: The Spine. Vol. 1, pp. 283-287. Philadelphia, W. B. Saunders, 1975.22. SHANNON, D. C.; RISEBOROUGH,E. J.; and KAZEMI, HOMAYOUN:Ventilation Perfusion Relationships Following Correction of Kyphoscoliosis.

J. Am. Med. Assn., 217: 579-584, 1971.23. TAKAHASHI, K.: Pulmonary Function in Scoliosis. J. Hokkaido Seikei-saigai Geka. 8: 17-39, 1963.24. TING, E. Y., and LYONS, H. A.: The Relation of Pressure and Volume of the Total Respiratory System and its Components in Kyphoscoliosis.

Am. Rev. Respir. Dis.. 89: 379-386, 1964.25. WESTIGATE,H. D., and MOE, J. H.: Pulmonary Function in Kyphoscoliosis Before and After Correction by Harrington Instrumentation Method.

J. Bone and Joint Surg., 51-A: 935-946, July 1969.26. ZORAB, P. A.; PRIME, F. J.; and HARRISON, AILIE: Lung Function in Young Persons after Spinal Fusion for Scoliosis. Spine, 4: 22-28, 1979.

THE JOURNAL OF BONE AND JOINT SURGERY on October 15, 2006 www.ejbjs.orgDownloaded from