and Respiratory Distress Syndrome

Transcription

1 Patent Ductus Arteriosus Ligation and Respiratory Distress Syndrome in Premature Infants David R. Clarke, M.D., Bruce C. Paton, M.D., Gary L. Way, M.D., and James R. Stewart, B.A. ABSTRACT Ligation of a patent ductus arteriosus was carried out in premature infants, 0 with concomitant respiratory distress. The duration of highvolume shunting is critical in determining the prognosis for these infants. Because of the low surgical mortality and morbidity and the high incidence of bronchopulmonary dysplasia in babies managed conservatively, infants with respiratory distress syndrome (RDS) who are respirator dependent should undergo ligation as soon as the presence of large left-to-right shunting is determined. Premature infants without RDS or those with mild RDS who are not respirator dependent can be managed medically or with elective ligation. Surgical intervention is strongly indicated in patients with persistent congestive heart failure and respiratory failure. Echocardiography offers an accurate and risk-free approach to the early diagnosis of a large left-to-right shunt through the ductus. Delayed spontaneous closure of the ductus arteriosus is common in premature infants. In infants without complicating respiratory problems or with mild respiratory distress syndrome (RDS) not requiring respirator support or positive-pressure ventilation, left-to-right shunting may persist for long periods with few deleterious effects. Among infants with severe RDS, however, mortality and morbidity rates are high, largely due to delayed pulmonary complications (ie, bronchopulmonary dysplasia). If the ductus arteriosus remains patent, or becomes patent, during an early hypoxic episode, respirator support may have to be continued long after primary pulmonary disease From the Departments of Cardiac Surgery and Pediatric Cardiology, University of Colorado Medical Center, Denver, co. Accepted for publication Feb, 19. Address reprint requests to Dr. Paton, Department of Cardiac Surgery, UniversityofColoradoMedicalCenter,00E Ninth Ave, Denver, CO 00. has cleared. In this situation the incidence of pulmonary complications due to prolongation of respirator support increases. Gay and co-workers [1 have defined large shunt duration (LSD) as the period from the onset of large-volume shunting (OLS) to the day of operation. Examination of our own patients and the relevant literature further establishes the value and significance of the LSD. The OLS has been defined clinically as the appearance of a typical murmur, bounding pulses, increased heart size, and greater pulmonary vascularity. Clinical criteria are unreliable as indicators of the presence of a large left-to-right shunt through a patent ductus arteriosus (PDA). The echocardiogram may offer a safe and effective method for confirming the magnitude of the shunt and early determination of the OLS. Material and Methods Twenty-two premature infants underwent extrapleural ligation of a PDA at the University of Colorado Medical Center between September, 19, and July, 19. Birth weights ranged from to,0 gm, and at the time of operation they varied from 0 to 3,0 gm (Table 1). All patients were operated upon through a left extrapleural posterolateral thoracotomy. In a few patients minimal tears were made in the pleura, but in no case was a chest tube left in place at the end of operation. The extrapleural approach was used to minimize handling of the lung and to avoid postoperative chest tube drainage. The average duration of operation was about minutes. Blood loss was negligible, and transfusion was never required. Twenty infants developed RDS of varying severity early in life. When respiratory failure occurred, continuous positive airway pressure (CPAP) or mechanical ventilation was used. All 0 infants were mechanically ventilated prior to 13

2 139 Clarke et al: FDA Ligation and Respiratory Distress Syndrome Table 1. Characteristics of the Study Population Birth Weight at Age at Patient Weight OLS LSD Operation Operation No. (gm) (days) (days) (gm) (days) Result 1 3 a " ,00 1,0 1,0 1,0,0 90 1, ,0 1,0 1,000 1, 1,0 90 1,0 1,00 1,000 3 wk Birth wk , , 3,0 1,03 0 1, ,30 1,10, 1,0 1, 00 1,00 1, "Twins. OLS = Onset of large-volume shunting; LSD = large shunt duration operation. The patients were divided into groups according to the severity of their RDS: infants without RDS ( patients), infants with mild RDS who were not respirator dependent ( patients), and infants with severe RDS who required either respirator assistance or CPAP (1 patients). In all cases the OLS was determined clinically as the onset of a typical murmur, bounding pulses, cardiomegaly, increased pulmonary vascularity, and rising Pa,-O,. The LSD was the interval between OLS and time of operation. Silverman and associates [1] first described the usefulness of the ratio between left atrial and aortic root diameters, as measured echocardiographically, in determining the magnitude of left-to-right shunting through a ductus. Echocardiograms were made in of the infants reported here, and the left atriallaortic root ratios were calculated. A ratio greater than 1.0 was taken as evidence of left-to-right shunting. Whenever possible, serial readings were made preoperatively, as persistently large or increasing ratios are more important than single estimations (Fig 1). Results The postoperative mortality rate (death during same hospital stay) was.%, of the infants. All patients who died had had RDS and were respirator dependent. One patient (No. 3) died days after going home; this was 9 days after operation and days after birth. The deaths in the hospital occurred,,,1, and days after operation, respectively; the causes of death are listed in Table. No infant died as a direct result of the operative procedure. Even early postoperative deaths were due to other causes such as necrotizing enterocolitis, contralateral respirator-induced pneumothorax, and central nervous system hemorrhage. Atelectasis and pneumothorax were observed

3 The Annals of Thoracic Surgery Vol No August 19 A Fig 1. (A) Preoperative and (B) postoperative echocardiograms showing reduction in left atriallaortic ratio. The postoperative echocardiogram was made hours after operation. (LA = left atrium.) both preoperatively and postoperatively but could not be attributed to the surgical procedure. Chronic respiratory disease persisted in 1 patient for two years postoperatively. The child was readmitted three times for pneumonia, asthma, and bronchitis. This complication was probably a result of prolonged respirator dependence. No instances of recurrent laryngeal neuropathy, pulmonary artery laceration, or other specifically surgical complications were observed. * Average FiO concentrations in patients who survived without bronchopulmonary dysplasia (BPD) were: 3% before operation, % the day after operation, % one week later. In infants who developed BPD, preoperative F I concentrations ~ ~ were greater than 0% and remained so postoperatively. B The OLS in the infants who died in the hospital occurred at a mean age of.0 days. The mean LSD for these patients was 1. days (f.) and ranged from 1 to days (Table 3). In infants who survived but who developed respirator-dependent RDS, the average OLS was.0 days. The mean LSD for these infants was.0 days (k.3). The LSD for infants with severe RDS who survived was never longer than that for similar infants who died. No deaths occurred among infants with mild RDS or without RDS, even though the LSD in these groups averaged 0. (f19.0) and 9. days (f1.), respectively. For infants with respirator-dependent RDS, the mortality rate associated with an LSD of days or less was 0. The mortality rate accompanying an LSD greater than days was 0%. Eleven infants underwent echocardiography Table. Causes of the Hospital Deaths Patient No. Age at death (days) Days Postop Cause of Death 1 BPD, RDS, left pneumothorax, CNS hemorrhage 9 BPD, 0, toxicity NEC, renal failure 1 0 Right pneumothorax 1 Unknown BPD = bronchopulmonary dysplasia; RDS = respiratory distress syndrome; NEC = necrotizing enterocolitis.

4 Clarke et al: PDA Ligation and Respiratory Distress Syndrome Table 3. Large Shunt Duration and Mortality Data in the Infants Severe RDS (N = 1) No RDS (N = ) Mild RDS (N = ) LSD in LSD in LSD in Survivors LSD in Survivors Survivors Nonsurvivors (days) (days) (days) (days) (1."; n = ) "Mean value for hospital mortality. LSD = large shunt duration. preoperatively. Accurate information was available for of the infants. The left atriallaortic root ratios varied between 1.0 and 1.3 (Table). Comment There is a very high incidence of patency of the ductus arteriosus after birth in premature infants. Rudolph [1] reported that nearly twothirds of all infants with a birth weight less than 1,0 gm have clinical evidence of a patent ductus, and in infants weighing less than 1,000 gm at birth the ductus is almost invariably patent. In a large percentage of premature infants, hyaline membrane disease is also present. Although there does not appear to be a direct causal relationship between patency of the ductus and RDS, both problems are related to prematurity. The high incidence of PDA in premature infants is probably due to both a high threshold of response of the immature ductus to oxygen and contractile inadequacy of the immature ductal musculature. An explanation for the apparently increasing clinical experience with PDA is that more preterm infants with RDS are surviving than was the case a few years ago. Paradoxically, the maintenance of Paoz below 0 mm Hg to avoid retrolental fibroplasia may further contribute to the high incidence of PDA. Presently, every effort is made to prevent the Pao, from rising above 0 to 0 mm Hg, a level that is not adequate to stimulate ductal constriction in premature infants. The successful treatment of RDS and prevention of retrolental fibroplasia may therefore combine to increase the incidence of PDA in these infants [. Early ligation of the PDA may or may not be indicated in preterm infants. Patent ductus ligation has been reported in 1 premature infants in the literature [,, 1. Although increasing numbers of infants have undergone ligation with varying results, the criteria for advocating operation are currently vague. Kilman and colleagues [] believe that Pace, is probably the most sensitive indicator of patient progress. If the Pace, increases in spite of ventilator support, the ductus should probably be ligated. Thibeault and co-workers [1] use the criteria of the need for increasing the inspired oxygen fraction and peak inspiratory pressure as well as heart size as indicators for surgical intervention. They also confirm the presence of left-to-right shunting through the ductus by retrograde aortography and determine the magnitude of shunting by the rate of disappearance of contrast medium from the aortic arch into the pulmonary artery. Coran and associates [31 use a figure of 0% shunt by indicator-dilution as a criterion for operation. These phenomena used to evaluate patient progress are consequences of a large left-to-right shunt. A more desirable approach would be to

5 The Annals of Thoracic Surgery Vol No August 19 Table. Echocardiographic Findings in Infants Interval from Interval from Ductus Size OLS to EC us to or Patient No. LNAO (mm) (days) (days) Large a.b 1. Large Large OLS btwins. LNAO = left atriudaorta ratio. OLS = onset of large shunting; EC = time of echocardiography; US = ultrasonography; OP = time of operation. attempt early detection of shunting by noninvasive means and to use this information, as well as the knowledge of the complications that may arise due to its presence, as criteria for advocating operation. Gay and colleagues [1 reported patients in whom the LSD was used as an indication for operation; it could also be correlated with the later development of bronchopulmonary dysplasia and was therefore of prognostic value. The LSD in their patients with severe RDS was 1. days in survivors and.1 days in nonsurvivors. The LSD in our patients with RDS who survived averaged.0 days, which suggests that the patients reported by Gay s group were operated on sooner than were ours. If the LSD criteria are applied to other reported series (Table ), the LSD for patients with severe RDS who survived was significantly shorter than for those who died (p < 0.01). Of 91 patients with RDS, lived and 33 died. Those who survived had a mean LSD of. days (+9.), whereas the LSD for those who did not live averaged.9 days (kll.l) (p < 0.). In these series, 1 patients had severe RDS; 31 lived and died. The average LSD for the survivors was 3.9 days (k.0), compared with 9.0 days (k.) for those who died. The average LSD among surviving infants with mild RDS was.3 days (k.); the average LSD among the infants with mild RDS who died was. days (f1.1) (p < 0.). These figures support the concept that the duration of the large shunt is critical in determining the mortality rate for infants with respirator-dependent RDS and PDA (Fig ). Current information indicates that for patients with RDS and a PDA who are respirator dependent, the shorter the LSD, the lower the mortality rate (Table ). Among reported patients there has been only l survivor in this group in whom the LSD was greater than days. In severe RDS, the mortality rate has been 9% in patients with an LSD greater than days but only 3.% when the LSD was less than days. The mortality rate in infants who underwent PDA ligation is summarized in Table. The

6 ~~ ~ ~~ ~ 13 Clarke et al: PDA Ligation and Respiratory Distress Syndrome Table. Large Shunt Duration and Reported Success in the Literature No RDS Mild RDS Severe RDS LSD LSD LSD LSD LSD LSD Author (days) L (days) D (days) L (days) D (days) L (days) D Clarke" Horsley" Gay" Murray" Neala Lewis" Kitterman Gupta" Thibeault 9. n =. n = n= n= 0. n = 1 n=. n = 3. n = 3 1 n = l 3. n = 1 n=. n = 3 n= n=l n= n=l n= "Used to calculate values quoted in text. RDS = respiratory distress syndrome; LSD = large shunt duration; L = lived; D = died..0 n = 1. n = 1 3. n = 0. n = 3 n=3 n=l. n = 1. n = n=3.1 n = 1. n = 3. n = n=l n= n=l. n = range is from 0% ill1 to 0% [9, 1. The mean mortality was %. Duration of medical therapy varied from 1 day to 3 weeks. None of the deaths Table. Large Shunt Duration and Hospital Mortality after PDA Ligation in Premature Infants with RDS were directly related to operation, and only occurred within hours of the procedure. Most Mortality (%) deaths were due to progressive respiratory failure 1 week to months followin? operation. Condition LSD < Days LSD > Days Other deaths were caused by necrotizing en- Severe RDS 3. 9 terocolitis, central nervous system hemorrhage, with PDA and meningitis. Most infants who died had RDS.3 1. bronchopulmonary dysplasia at the time of their with PDA demise, regardless of the cause of death. Mortality rates, however, do not tell the entire story. There are patients who survive but remain respiratory cripples because of bronchopulmonary dysplasia. One infant in our Table. Hospital Mortality after PDA Ligation in Premature Infants Deaths No. of Author Patients Number O/O LSD in days Fig. Survival and death at different shunt durations among patients with severe respiratory distress syndrome. (LSD = large shunt duration.) Kilman Horde y Murphy Gupta Ki tterman Gay Lewis Neal Thibeault Coran Clarke 3 Total 1

7 1 The Annals of Thoracic Surgery Vol No August 19 series, followed for two years, had numerous admissions related to bronchitis, pneumonia, and asthma. Thus, there may be chronic morbidity primarily due to prolonged respirator support during early infancy. In all series the OLS was determined from the clinical criteria presented in the respective study. If detailed information as described earlier was not available, relevant information from the report (eg, onset of murmur, congestive heart failure, or cardiomegaly) was used to estimate the OLS. The LSD was determined in the usual manner. Echocardiographic measurement of the left atriayaortic ratio can be used to detect a significant left-to-right shunt prior to onset of many of the typical physical findings. In 19 Silverman and co-workers [1] reported utilizing the ratio as an indication for operation. Baylen and associates [l] have described the usefulness of the left atrial and left ventricular end-diastolic diameters for detecting large left-to-right shunts. The left atriayaortic ratio was found to be very important in assessing the need for operation in these infants, especially when the presence of RDS made radiographic estimation of heart size difficult. It is possible to make an echocardiographic diagnosis of PDA hours before physical findings make the diagnosis clear; the technique is useful in assessing the efficacy of treatment for congestive failure. There is good correlation between the left atriayaortic ratio and the size of the ductus as measured during operation. In the patients in whom the ratio was less than 1.0, the ductus was found to be small and probably was not contributing to the patient s problems. After the PDA is ligated the ratio should return rapidly to normal, sometimes within an hour after operation. Conclusions Mortality rates for newborn infants with RDS (hyaline membrane disease) are to 0% [, 131. Death is usually a result of oxygen toxicity and severe pulmonary disease. It is well known that premature infants who require mechanical ventilation and subsequently develop large shunting through a PDA have a poor prognosis (9% mortality in this study) and frequently die of complications related to prolonged respirator support. Early surgical ligation of a large PDA in infants with respirator-dependent RDS is critical if these patients are to survive with normal lungs. Many clinicians take the attitude that infants with severe RDS are too sick to undergo surgical ligation, and choose medical therapy for varying lengths of time after the clinical OLS is observed. Indeed, the infant is too ill to forego operative closure of the ductus by this time. The operative risks reported using the extrapleural approach are minimal in nearly every series (none of the deaths in the studies examined were due to surgical complications), and others have reported equally excellent results with the transpleural method of ductal ligation. Hospital mortality with either technique for closure of uncomplicated PDA by experienced hands is essentially zero [1. Echocardiography-specifically the left atrial/ aortic root ratio-offers an accurate, inexpensive method for early determination of the OLS, before a murmur, bounding pulses, cardiomegaly, and increased pulmonary vascularity are present. Infants with mild RDS or without RDS who develop a PDA can be managed medically until spontaneous closure occurs, elective operation is chosen, or operation becomes imperative due to persistent congestive heart failure and impending respiratory failure. If the severity of the RDS is in question or if the left atriayaortic ratio is 1.0 to 1., medical treatment can safely be followed for hours before deciding about the need for operation. References 1. Baylen BG, Meyer RA, Kaplan S, et al: Echocardiographic assessment of severity of patent ductus arteriosus with pulmonary disease. J Pediatr :3, 19. Behrman RE: Neonatology: Diseases of the Fetus and Infant. St Louis, Mosby, Coran AG, Cabal L, Siassi, et al: Surgical closure of patent ductus arteriosus in the premature infant with respiratory distress. J Pediatr Surg :399, 19. Gay JH, Daily WJR, Meyer BHP, et al: Ligation of

8 1 Clarke et al: FDA Ligation and Respiratory Distress Syndrome the patent ductus arteriosus in premature infants: report of cases. J Pediatr Surg :,193. Gupta JM, Fisk GC, Wright JS: Ductus ligation in respiratory distress syndrome. J Thorac Cardiovasc Surg 3:, 19. Horsley BL, Lerberg DB, Allen AC, et al: Respiratory distress from patent ductus arteriosus in the premature newborn. Ann Surg 1:0, 193. Hurst JW: The Heart. Third edition. New York, Blakiston Div, McGraw-Hill, 19. Kilman JW, Kakos GS, Williams TE Jr, et al: Ligation of patent ductus arteriosus for persistent respiratory distress syndrome in premature infants. J Pediatr Surg 9:, Kitterman JA, Edmunds H Jr, Gregory GA, et al: Patent ductus arteriosus in premature infants: incidence, relation to pulmonary disease and management. N Engl J Med :3, 19. Lewis CE Jr, Coen RW, Talbot W, et al: Early intervention in premature infants with respira- tory distress syndrome and patent ductus arteriosus. Am J Surg 39, 19. Murphy DA, Outerbridge E, Stern L, et al: Management of premature infants with patent ductus arteriosus. J Thorac Cardiovasc Surg :1,19. Neal WA, Bessinger FB Jr, Hunt CE, et al: Patent ductus arteriosus complicating respiratory distress syndrome. J Pediatr :, Robbins SLR: Pathologic Basis of Disease. Philadelphia, Saunders, Rudolph AM: Congenital Diseases of the Heart. Chicago, Year Book, Silverman NH, Lewis AB, Heymann MA, et al: Echocardiographic assessment of ductus arteriosus shunt in premature infants. Circulation 0:1, Thibeault DW, Emmanouilides GC, Nelson RJ, et al: Patent ductus arteriosus complicating the respiratory distress syndrome in preterm infants. J Pediatr :0, 19 Notice from the Southern Thoracic Surgical Association The Twenty-third Annual Meeting of the South- idents, authors in the program, invited speakern Thoracic Surgical Association will be held at ers, and thoracic surgeons from Mexico. Do not the Princess Hotel, Acapulco, Mexico, on No- write to the hotel for reservations. Applications vember -, 19. The scientific program will for room and group plane reservations may be be published in the September issue of The An- obtained from: nals of Thoracic Surgery. Applications for membership must be corn- American Express pleted by September 1, 19, and forwarded to STSA Special Desk Orin D. Butterick, M.D., 9 Madison Ave, 13 Carondelet St Memphis, TN 33. New Orleans, LA 01 Nonmember physicians are welcome to attend the meeting. There will be a $0 registra- I. Kent Trinkle, M.D. tion fee for nonmembers excluding interns, res- Secretary-Treasurer