- Email: [email protected]
Congenital adenomatoid malformation of the lung: Current management and prognosis
Congenital Adenomatoid Malformation of the Lung: Current Management and Prognosis ByI.R.Neilson,
P. Russo, J.-M. Laberge,
D. Filiatrault,
L.T. Nguyen,
P.P. Collin, and F.M. Guttman
Montreal, Quebec 0 A retrospective lo-year review of congenital adenomatoid malformation (CAM) included 10 cases diagnosed in utero by ultrasound and 13 cases that presented postnatally. Two prenatally diagnosed cases were aborted because of associated lethal anomalies. All remaining patients underwent resection. Up to one third of all cases, whether diagnosed prenatally or postnatally, were asymptomatic. Resection is recommended to avoid respiratory distress, infection, or associated malignancy. There were 5 nonsurvivors, including 2 therapeutic abortions and 3 who died postoperatively. All had either polyhydramnios or ascites. All patients who died postoperatively had a respiratory arrest at birth and underwent immediate lobectomy. All died on the first day of life after a brief period and were found to have associated pulmonary hypoplasia. One had undergone a prenatal transthoracic cyst aspiration at 34 weeks gestation in an attempt to allow lung growth and prevent premature labor. Prenatal ultrasound findings suggestive of poor prognosis included polyhydramnios, ascites, mediastinal shift, and noncystic type Ill CAM. However, there is a spectrum of severity of CAM. The lesion can either regress and be asymptomatic at birth, or it can progress to cause either fetal death from hydrops or neonatal death caused by associated pulmonary hypoplasia. These findings should be considered in prenatal counseling for CAM Copyright o 1991 by WA Saunders Company INDEX WORDS: Congenital cystic adenomatoid tion; lung cyst; fetal diagnosis and treatment.
malforma-
ONGENITAL ADENOMATOID malformation (CAM) of the lung is both a rare developc mental lesion and an unusual cause of respiratory distress at birth. CAM has been classified by Stocker, Madewell, and Drake’ according to the presence and size of the cystic component. Type I lesions are composed of single or multiple cysts, some of which are larger than 1 cm in diameter. Type II lesions are intermediate with uniform cysts, all less than 1 cm. Type III lesions appear solid and are composed of bronchiolar-like structures without alveolar differentiation. This entity was first distinguished from other types of congenital cystic lung disease in 1949 by Ch’in and Tang.* These lesions are relatively rare; a total of only 421 cases was compiled in a recent international review.j However, with the introduction of routine prenatal ultrasound examination, there has been a rapid increase in the number of cases detected. Because there seems to be a correlation between the morphologic classification and the natural history of this disease,4 ultrasound detection may influence the prenatal as well as postnatal management of this JournalofPediafric Surgery, Vol26, No 8 (August), 1991: pp 975-981
disease. A retrospective review of all patients with CAM was carried out to correlate the clinical and pathological findings with outcome. MATERIALS
AND METHODS
Medical records of patients with a CAM diagnosed over a lo-year period ending June 1990 at the Montreal Children’s Hospital and Hospital Sainte-Justine in Montreal, Canada were reviewed. RESULTS
There were 23 patients; 17 boys and 6 girls. In 10 cases, the lesion was diagnosed by prenatal ultrasound. In the other 13 cases, diagnosis was made postnatally; it could not be determined whether a prenatal ultrasound had been performed. Two additional patients with a presumed diagnosis of CAM who have not yet undergone operation are not included in this series. Prenatal Presentation (10 Cases)
The mean gestational age of prenatal diagnosis by obstetrical ultrasound in 10 cases (Table 1) was 26.4 weeks (range, 18 to 36 weeks). Polyhydramnios was noted in 5 cases of which only 2 survived. However, both survivors had only transient polyhydramnios. The CAM was shown to regress in utero in case 2. Ascites was detected in 4 cases, only one of whom survived. Two of these fetuses also had multiple life-threatening anomalies and were aborted. Eight cases were followed-up to delivery, with one (case 8) undergoing thoracic cyst aspiration in utero. At birth the mean gestational age was 36.4 weeks (range, 30 weeks to term). Five (63%) had respiratory distress including 2 (25%) with respiratory arrest. Only three (37%) were asymptomatic. Eight patients have undergone lobectomy. Age at operation ranged from the first day of life to 3% months of age. Anomalies were From the Department of Pediatric Sutgety, Montreal Children’s Hospital, McGill University and the Hopital Sainte-Justine, Universite de Montreal, Montreal, Quebec. Presented at the Jens G. Rosenkrantz Resident Competition at the 42nd Annual Meeting of the Surgical Section of the American Academy of Pediatrics, Boston, Massachusetts, October 6-7, 1990. Address reprint requests to Jean-Martin Laberge MD, Department of Pediatric Surgery, Montreal Children’s Hospital, McGill University, 2300 Tupper St, Montreal, Canada H3H IP3. Copyright o 1991 by W.B. Saunders Company 0022-3468191/2608-0017$03.0010
975
NEILSON ET AL
co
M
=_
rm
=
8
---
-or
infrequent in survivors and included one bilateral cleft lip and palate and one previously reported case in which the CAM was associated with an extralobar sequestration.’ One additional patient was diagnosed with a suspected CAM at 18 weeks’ gestation. Due to a high degree of mediastinal shift, it was initially recommended to terminate pregnancy. However, the mediastinal shift decreased and the fetus was followed-up to term delivery. The child was asymptomatic at birth and so remains at 20 months of age while awaiting resection. Both the chest radiography (CXR) and computed tomography (CT) are consistent with a right lower lobe CAM. Postnatal Presentation (13 Cases)
Of the 13 cases diagnosed postnatally (Table 2), only one had a history of polyhydramnios at delivery. This only nonsurvivor had a resection of a type III CAM on the first day of life. None of the cases had any record of whether a prenatal ultrasound had been performed. The mean gestational age at birth was 37.3 weeks (range, 32 weeks to term). Onset of respiratory symptoms began during the neonatal period in only 6 cases (46%) with one having a respiratory arrest. Four cases (31%) were diagnosed from incidental CXR findings during investigation of unexplained fever. Only 3 (23%) were asymptomatic. Lobectomy was performed in all patients from the first day of life to 8 years of age, with 5 being over 1 year of age at the time of operation. Again, anomalies were rare, with only one associated pectus excavatum deformity. One additional patient not included in the series was delivered by cesarean section for fetal distress with meconium staining. There was mild respiratory distress at birth, but the patient became asymptomatic in the neonatal period and has been followed-up for 6 years without referral for operation. Radiologically, this lesion is typical of a CAM. Investigations
Findings on CXR included cysts (12), mediastinal shift (7), and the presence of an opacity (7). The initial CXR diagnosis was correct in 15 cases; erroneous diagnoses included congenital lobar emphysema (3), chronic atelectasis (l), pneumatocele (1) and undetermined (3). A CT scan was helpful in 6 cases. Other investigations such as ventilation perfusion scans (3), arteriograms (2), upper gastrointestinal series (2) and bronchoscopy (3) were helpful only occasionally. It should be noted that antenatal ultrasound diagnosis was correct in all cases, although the possibility of a congenital diaphragmatic hernia was briefly entertained in one case.
CONGENITAL
ADENOMATOID
977
MALFORMATION
Table 2. Summary of Cases Diagnosed Poetnatally Presentation CSS.S
GA Iwk)
11
Complications
Birth
34
Outcome
Operation
Description
AaS
TVPS
Alive
Moderate respiratory distress, tachypnea
3mo
111
1 vr
Anomalies
Dead
recurred at 3 mo 12
36
7Wk
Asymptomatic; UTI workup
3mo
13
32
Diabetes
Birth
Severe respiratory distress, arrest
Id
14
37
Twin
15
40
4 vr 1.5mo
16
40
10d
17
40
18
40
I
2mo
III
Operative
6 yr 15 mo
I
3mo
I
4
Mild respiratory distress; Fever workup
1 mo
I
Asymptomatic Asymptomatic
vr
6vr
R lung abscess/pneumatocele
6 yr
Ii
3 vr 3mo
Pectus
1 mo
Resection CAM about 3 mo
5 yr
II
1 yr
Excavatum
*5 yr
Mild respiratory distress Fever workup
8 yrs
II
1 yr
Mild respiratory distress
2mo
I
19
38
20
40
8 vr 7wk
21
28
Birth
Mild respiratory distress
1 mo
I
22
40
Birth
Mild respiratory distress
II
23
40
6mo
Mild respiratory distress
2 yr 7mo
I
*Incomplete resection done elsewhere at 3 months of age, presented with recurrent respiratory distress and pectus excavatum.
Pathology
Case 3
Lesions were classified (according to Stocker) as follows: 13 with type I (57%), 6 with type II (26%), and 4 with type III (17%) lesions. The right lung was affected twice as frequently as the left, with lesions distributed equally among all lobes. The classification in nonsurvivors were one type I, one type II, and three type III lesions. The two aborted fetuses had type III lesions, one had total bilateral and the other had whole right lung involvement. Overall, only one patient with a type III lesion has survived.
A diagnosis of CAM was made at 30 weeks’ gestation. Other findings included polyhydramnios and ascites. An emergency cesarean section was performed for fetal distress. A hydropic male (Fig 3) was born with Apgars of 2 and 3 at I and 5 minutes, respectively. A left lower lobectomy was immediately performed and a 78-g type II CAM resected. The patient died at 22 hours of life. An autopsy demonstrated associated pulmonary hypoplasia.
Outcome The delay between presentation of symptoms and age at operation was less than 1 week in 7 cases, within 1 year in 8 cases, and more than 1 year in 6 cases. The three patients who died within the first 24 hours postoperatively with documented pulmonary hypoplasia as well as a CAM underwent resection immediately after birth for severe respiratory distress. CASE REPORTS
Case 8 Obstetrical ultrasound showed a fetus having two large thoracic cysts and polyhydramnios at 30 weeks’ gestation, compatible with either CAM or congenital diaphragmatic hernia. Fetal echocardiography was normal and no other anomalies were detected. While fetal intervention was being considered, premature rupture of membranes occured at 34 weeks’ gestation. Repeat ultrasound clearly demonstrated a CAM with two large cysts and mediastinal shift (Figs 4 and 5). Cyst aspiration was attempted to try to lessen medistinal shift, to reduce lung compression, and to relieve polyhydramnios. Because of fetal position and movement, only one cyst could be aspirated and it was not possible to insert “pig-tail” catheters. Spontaneous delivery occurred 2 days later with severe perinatal respiratory distress. At 3 hours of life, a left pneumonectomy was performed. Transient improvement in oxygen saturation from 70% to 90% on 100% oxygen was followed by a rapid deterioration and the patient died at 8 hours of life. The total left lung weight was 38 g with the upper lobe having pulmonary hypoplasia and the lower a type I CAM.
Case 2 An infant had a right thoracic Iesion consistent with a CAM diagnosed at 18 weeks’ gestation. The mass caused mediastinal shift but there was no polyhydramnios, ascites, or other signs of hydrops. Possible regression was taken into account in prenatal counseling. The lesion was followed-up and progressively decreased in size from 4.5 to 2.7 cm (Fig 1) by 30 weeks. After term delivery, there were no respiratory symptoms and the CXR was almost normal (Fig 2). A CXR repeated at 3 months of age showed some mass effect, but the infant continued to be asymptomatic. An elective lobectomy was performed and a 31-g right lower lobe type I CAM was removed.
Case 14 A boy first presented at 3 years of age with an episode of croup. On examination there were decreased breath sounds and an end expiratory wheeze in the left upper chest. A CXR showed a cystic lesion in the left upper lobe that was interpreted as congenital lobar emphysema (Fig 6). Bronchoscopy was consistent with this diagnosis and he was followed-up for 2 years without any symptoms or change in the CXR. However, a CT scan was consistent with a CAM (Fig 7), so a left upper lobectomy was performed which confirmed a type I CAM.
978
NEILSON ET AL
Fig 1. Fetal ultrasound of type I CAM in case 2. On the longitudinal view (A) taken at 18 weeks are arrows overlying the fetal spine and showing the 4.5-cm mass. The transverse view(B) taken at 27 weeks shows the right thoracic mass has decreased to 3.0 cm, but still shifts the heart toward the left thorax.
DISCUSSION
This series was collected over a period during which routine prenatal ultrasound screening became widely available. Compared with most previous reports,3,6,7it is a large series and represents a relatively homogeneous population because it encompasses only a IO-year period. Ultrasound scanning correctly diagnosed CAM in the 10 prenatal cases who either underwent resection or therapeutic abortion. Ultrasound was helpful in recognizing neonates at high risk for perinatal respiratory distress. Prenatal diagnosis permitted transfer to
Fig 2.
CXR of case 2 showing faint cysts in right lower lobe.
a tertiary center for high risk obstetrical and neonatal care. As a result, all symptomatic neonates were operated on promptly. In contrast, those who presented in the postnatal period did not have the benefit of prompt recognition. Approximately 40% of these neonates had respiratory symptoms at birth. However, this delay did not appear to affect morbidity or mortality. The prenatal factors associated with a poor outcome were the detection of polyhydramnios, ascites,
Fig 3. Radiograph of cese 3 showing left lower lobe type II CAM, mediastinal shift, and ascites.
CONGENITAL ADENOMATOID
979
MALFORMATION
Fig 4. Fetal ultrasound of case 8 showing 5.2-cm type I CAM before cyst aspiration at 33 weeks’ gestation.
or mediastinal shift. As well, the more hyperechoic type III CAM, which has a much worse prognosis, could be differentiated from cystic forms of CAM by fetal ultrasound. Ultrasound also was able to detect two fetuses with multiple life-threatening anomalies. These were both associated with type III CAM lesion as well as polyhydramnios and ascites. However, it is important not to overestimate the poor prognosis of a prenatally detected lesion because there were two survivors with polyhydramnios that was only transient. Likewise, in one case, the CAM progressively decreased in size; in another case, associated mediastinal shift disappeared. Partial regression in utero of CAM has also been documented in two case reports.’ The three postnatal deaths in this series were associated with perinatal respiratory arrest with emergency resection undertaken as a last resort. All three
Fig 5. CXR of case 8 at birth showing left lower lobe CAM and mediastinal shift.
Fig 6. Preoperative CXR of case 14 erroneously congenital lobar emphysema.
interpreted
as
died within the first postoperative day, often with a brief postoperative honeymoon period during which respiratory parameters improved. All were later shown to have pulmonary hypoplasia secondary to the CAM. Drainage of large space occupying cysts to reverse fetal hydrops is physiologically sound. Repeat needle aspirations or insertion of in utero thoracoamniotic shunts have been described, occasionally with successful outcome.9-11 Fetal “chest tubes” have also been found to reverse polyhydramnios and hydrops in fetal hydrothorax, which also acts as a space occupying lesion.“,” The management of solid type III CAM in the fetus would be more difficult, requiring open fetal surgical intervention as reported by Harrison et al for diaphragmatic hemia.14 Some cases with less severe
Flg 7. Chest CT scan of case 14 documenting left upper lobe CAM with a large unilocular cyst and several smaller cysts anteriorly and adjacent to the mediastinum.
980
NEILSON ET AL
lung hypoplasia may also benefit from perioperative extracorporal membrane oxygenation support. In contrast to those who died, all infants without a preoperative respiratory arrest have survived. There was one survivor with a type III lesion, which is usually associated with a fatal outcome. Adzick et al4 has reported one other survivor with a type III lesion, In the antenatal diagnosis group, one half of the survivors remained asymptomatic in the newborn period; hence, these were really incidentally diagnosed lesions. However, we would recommend prompt surgery after diagnosis of CAM in all cases. In this series, one infant is awaiting resection of a presumed CAM and another 6-year-old patient has been followed-up from birth with a suspected CAM. Because of the reports of malignancy associated with CAM in young adults,‘S-‘8 as well as the risk of pulmonary
infections, we feel it is important to resect all suspected CAM lesions. In summary, CAh4 has a spectrum of severity, and even when diagnosed in utero it can either regress and be asymptomatic at birth, or can progress to cause either fetal death or neonatal respiratory arrest from pulmonary hypoplasia. Serial ultrasound examinations are required to assess prognosis and allow counseling; however, polyhydramnios and fetal ascites require some form of intervention if death is to be avoided.
ACKNOWLEDGMENT The authors thank Dr Gisele Lussier, radiologist, who provided the fetal ultrasounds in case 2 and was instrumental in the referral for prenatal counseling.
REFERENCES 1. Stocker JT, Madewell JE, Drake RM: Congenital cystic adenomatoid malformation of the lung: Classification and morphologic spectrum. Hum Path01 8:155-171,1977 2. Ch’in KY, Tang MY: Congenital adenomatoid malformation of one lobe of a lung with general anasarca. Arch Path01 48:221229,1949 3. Beluffi G, Brokensha C, Kozlowski K, et al: Congenital cystic adenomatoid malformation of the lung. Presentation of 16 cases. Fortschr Rontgenstr 150:523-530,1989 4. Adzick NS, Harrison MR, Glick PL, et al: Fetal cystic adenomatoid malformation: Prenatal diagnosis and natural history. J Pediatr Surg 20:483-4881985 5. Morin C, Filiatrault D, Russo P: Pulmonary sequestration with histologic changes of cystic adenomatoid malformation. Pediatr Radio1 18:130-l’32,1989 6. Ostor AG, Fortune DW: Congenital cystic adenomatoid malformation of the lung. Am J Clin Path01 70:595-604,1978 7. Wesley JR, Heidelberger KP, DiPietro MA, et al: Diagnosis and management of congenital cystic disease of the lung in children. J Pediatr Surg 21:202-207,1986 8. Saltzman DH, Adzick NS, Benacerraf BR: Fetal cystic adenomatoid malformation of the lung: Apparent improvement in utero. Obstet Gynecol71:1000-1002,1988 9. Chao A, Monoson RF: Neonatal death despite fetal therapy for cystic adenomatoid malformation: A case report. J Reprod Med 35:655-657,199O
10. Nugent CE, Hayashi RH, Rubin J: Prenatal treatment of type I congenital cystic adenomatoid malformation by intrauterine fetal thoracentesis. J Clin Ultrasound 17:675-677,1989 11. Clark SL, Vitale DJ, Minton SD, et al: Successful fetal therapy for cystic adenomatoid malformation associated with second-trimester hydrops. Am J Obstet Gynecol157:294-295,1987 12. Longaker MT, Laberge J-M, Dansereau J, et al: Primary fetal hydrothorax: Natural history and management. J Pediatr Surg 24:573-576,1989 13. Rodeck CH, Fisk NM, Fraser DI, et al: Long-term in utero drainage of fetal hydrothorax. N Engl J Med 319:1135-1138,1988 14. Harrison MR, Adzick NS, Longaker MT, et al: Successful repair in utero of a fetal diaphragmatic hernia after removal of herniated viscera from the let thorax. N Engl J Med 3221582-1584, 1990 15. Sheffield EA, Addis BJ, Corrin B, et al: Epithelial hyperplasia and malignant change in congenital lung cysts. J Clin Path01 40:612-614,1987 16. Prichard MG, Brown PJE, Sterrett GF: Bronchioloalveolar carcinoma arising in longstanding lung cysts. Thorax 39545-549, 1984 17. Weinberg AG, Currarino G, Moore GC, et al: Mesenchymal neoplasia and congenital pulmonary cysts. Pediatr Radio1 9:179182,198O 18. Weinblatt ME, Siegel SE, Isaacs H: Pulmonary blastoma associated with cystic lung disease. Cancer 49:669-671,1982
Discussion A. Hafler (Baltimore, MD): Dr Nielson, I listened to your report with great interest. I kept waiting for you to give us the punchline, which your abstract says is “to find out whether prenatal diagnosis would influence the management of this lesion.” I hope you will indicate to us whether the prenatal studies have altered your management of the patients. I say this because your data show us clearly that one of the type
III patients survived. It also shows us that there can be marked changes even in the early prenatally diagnosed patient in the course of the further development of the lung. I urge caution in supporting abortion because I have had one patient I was quite certain would not survive with bilateral cystic lesions. Because the parents did not wish to interrupt the pregnancy, the patient was carried to term, was
CONGENITAL ADENOMATOID
MALFORMATION
delivered, and is now 8 years old. The patient did require initial resection on one side and, subsequently, has required resection of a small cystic area on the other. It seems to me that we must be very careful in interpreting this. I believe it is important to get a better understanding of the pathophysiology, but I find it difficult to try to use this in a retrospective way. I believe that we should allow these babies to tell us after they’re born whether they are capable of surviving. The concern I have is that the polyhydramnios is not necessarily as you’ve shown us: that is there’s going to be bilateral hypoplasia of the lungs that will not allow survival. I would also like to ask you about hydrops. Do you have a good explanation for that and is it a reversible component also? These studies are very important, but I think that we shouldn’t jump to conclusions until we have a better understanding of the pathophysiology of the lesions. LR. Neilson (response): I’d like to thank Dr Haller for prompting me to stress a very important part of this paper. This concerns the management of CAM once an in utero diagnosis is made. We studied which
981
factors were important in the outcome of CAM, in order to allow more effective prenatal counseling. In this series, there were only two cases that were terminated mainly for multiple associated lifethreatening anomalies, ie, renal agenesis, rather than for the CAM. All of the other fetuses were followed to term. Detection of a type III lesion is not an indication for termination, especially because there was one survivor in this series with a documented type III lesion. Of all prognostic factors, the size of the cyst probably is the most important, because this leads to shift of the mediastinum and indirectly influences the pathogenesis of both polyhydramnios and ascites. We attempted cyst aspiration in one fetus; however, this was probably done too late in pregnancy. There are several case reports in the literature of successful treatment of polyhydramnios, as well as the regression of fetal hydrops. All of the patients in this series who died postoperatively had documented pulmonary hypoplasia. Obviously, if progress is to be made in preventing mortality, fetal intervention will need to be further assessed.
P. Russo, J.-M. Laberge,
D. Filiatrault,
L.T. Nguyen,
P.P. Collin, and F.M. Guttman
Montreal, Quebec 0 A retrospective lo-year review of congenital adenomatoid malformation (CAM) included 10 cases diagnosed in utero by ultrasound and 13 cases that presented postnatally. Two prenatally diagnosed cases were aborted because of associated lethal anomalies. All remaining patients underwent resection. Up to one third of all cases, whether diagnosed prenatally or postnatally, were asymptomatic. Resection is recommended to avoid respiratory distress, infection, or associated malignancy. There were 5 nonsurvivors, including 2 therapeutic abortions and 3 who died postoperatively. All had either polyhydramnios or ascites. All patients who died postoperatively had a respiratory arrest at birth and underwent immediate lobectomy. All died on the first day of life after a brief period and were found to have associated pulmonary hypoplasia. One had undergone a prenatal transthoracic cyst aspiration at 34 weeks gestation in an attempt to allow lung growth and prevent premature labor. Prenatal ultrasound findings suggestive of poor prognosis included polyhydramnios, ascites, mediastinal shift, and noncystic type Ill CAM. However, there is a spectrum of severity of CAM. The lesion can either regress and be asymptomatic at birth, or it can progress to cause either fetal death from hydrops or neonatal death caused by associated pulmonary hypoplasia. These findings should be considered in prenatal counseling for CAM Copyright o 1991 by WA Saunders Company INDEX WORDS: Congenital cystic adenomatoid tion; lung cyst; fetal diagnosis and treatment.
malforma-
ONGENITAL ADENOMATOID malformation (CAM) of the lung is both a rare developc mental lesion and an unusual cause of respiratory distress at birth. CAM has been classified by Stocker, Madewell, and Drake’ according to the presence and size of the cystic component. Type I lesions are composed of single or multiple cysts, some of which are larger than 1 cm in diameter. Type II lesions are intermediate with uniform cysts, all less than 1 cm. Type III lesions appear solid and are composed of bronchiolar-like structures without alveolar differentiation. This entity was first distinguished from other types of congenital cystic lung disease in 1949 by Ch’in and Tang.* These lesions are relatively rare; a total of only 421 cases was compiled in a recent international review.j However, with the introduction of routine prenatal ultrasound examination, there has been a rapid increase in the number of cases detected. Because there seems to be a correlation between the morphologic classification and the natural history of this disease,4 ultrasound detection may influence the prenatal as well as postnatal management of this JournalofPediafric Surgery, Vol26, No 8 (August), 1991: pp 975-981
disease. A retrospective review of all patients with CAM was carried out to correlate the clinical and pathological findings with outcome. MATERIALS
AND METHODS
Medical records of patients with a CAM diagnosed over a lo-year period ending June 1990 at the Montreal Children’s Hospital and Hospital Sainte-Justine in Montreal, Canada were reviewed. RESULTS
There were 23 patients; 17 boys and 6 girls. In 10 cases, the lesion was diagnosed by prenatal ultrasound. In the other 13 cases, diagnosis was made postnatally; it could not be determined whether a prenatal ultrasound had been performed. Two additional patients with a presumed diagnosis of CAM who have not yet undergone operation are not included in this series. Prenatal Presentation (10 Cases)
The mean gestational age of prenatal diagnosis by obstetrical ultrasound in 10 cases (Table 1) was 26.4 weeks (range, 18 to 36 weeks). Polyhydramnios was noted in 5 cases of which only 2 survived. However, both survivors had only transient polyhydramnios. The CAM was shown to regress in utero in case 2. Ascites was detected in 4 cases, only one of whom survived. Two of these fetuses also had multiple life-threatening anomalies and were aborted. Eight cases were followed-up to delivery, with one (case 8) undergoing thoracic cyst aspiration in utero. At birth the mean gestational age was 36.4 weeks (range, 30 weeks to term). Five (63%) had respiratory distress including 2 (25%) with respiratory arrest. Only three (37%) were asymptomatic. Eight patients have undergone lobectomy. Age at operation ranged from the first day of life to 3% months of age. Anomalies were From the Department of Pediatric Sutgety, Montreal Children’s Hospital, McGill University and the Hopital Sainte-Justine, Universite de Montreal, Montreal, Quebec. Presented at the Jens G. Rosenkrantz Resident Competition at the 42nd Annual Meeting of the Surgical Section of the American Academy of Pediatrics, Boston, Massachusetts, October 6-7, 1990. Address reprint requests to Jean-Martin Laberge MD, Department of Pediatric Surgery, Montreal Children’s Hospital, McGill University, 2300 Tupper St, Montreal, Canada H3H IP3. Copyright o 1991 by W.B. Saunders Company 0022-3468191/2608-0017$03.0010
975
NEILSON ET AL
co
M
=_
rm
=
8
---
-or
infrequent in survivors and included one bilateral cleft lip and palate and one previously reported case in which the CAM was associated with an extralobar sequestration.’ One additional patient was diagnosed with a suspected CAM at 18 weeks’ gestation. Due to a high degree of mediastinal shift, it was initially recommended to terminate pregnancy. However, the mediastinal shift decreased and the fetus was followed-up to term delivery. The child was asymptomatic at birth and so remains at 20 months of age while awaiting resection. Both the chest radiography (CXR) and computed tomography (CT) are consistent with a right lower lobe CAM. Postnatal Presentation (13 Cases)
Of the 13 cases diagnosed postnatally (Table 2), only one had a history of polyhydramnios at delivery. This only nonsurvivor had a resection of a type III CAM on the first day of life. None of the cases had any record of whether a prenatal ultrasound had been performed. The mean gestational age at birth was 37.3 weeks (range, 32 weeks to term). Onset of respiratory symptoms began during the neonatal period in only 6 cases (46%) with one having a respiratory arrest. Four cases (31%) were diagnosed from incidental CXR findings during investigation of unexplained fever. Only 3 (23%) were asymptomatic. Lobectomy was performed in all patients from the first day of life to 8 years of age, with 5 being over 1 year of age at the time of operation. Again, anomalies were rare, with only one associated pectus excavatum deformity. One additional patient not included in the series was delivered by cesarean section for fetal distress with meconium staining. There was mild respiratory distress at birth, but the patient became asymptomatic in the neonatal period and has been followed-up for 6 years without referral for operation. Radiologically, this lesion is typical of a CAM. Investigations
Findings on CXR included cysts (12), mediastinal shift (7), and the presence of an opacity (7). The initial CXR diagnosis was correct in 15 cases; erroneous diagnoses included congenital lobar emphysema (3), chronic atelectasis (l), pneumatocele (1) and undetermined (3). A CT scan was helpful in 6 cases. Other investigations such as ventilation perfusion scans (3), arteriograms (2), upper gastrointestinal series (2) and bronchoscopy (3) were helpful only occasionally. It should be noted that antenatal ultrasound diagnosis was correct in all cases, although the possibility of a congenital diaphragmatic hernia was briefly entertained in one case.
CONGENITAL
ADENOMATOID
977
MALFORMATION
Table 2. Summary of Cases Diagnosed Poetnatally Presentation CSS.S
GA Iwk)
11
Complications
Birth
34
Outcome
Operation
Description
AaS
TVPS
Alive
Moderate respiratory distress, tachypnea
3mo
111
1 vr
Anomalies
Dead
recurred at 3 mo 12
36
7Wk
Asymptomatic; UTI workup
3mo
13
32
Diabetes
Birth
Severe respiratory distress, arrest
Id
14
37
Twin
15
40
4 vr 1.5mo
16
40
10d
17
40
18
40
I
2mo
III
Operative
6 yr 15 mo
I
3mo
I
4
Mild respiratory distress; Fever workup
1 mo
I
Asymptomatic Asymptomatic
vr
6vr
R lung abscess/pneumatocele
6 yr
Ii
3 vr 3mo
Pectus
1 mo
Resection CAM about 3 mo
5 yr
II
1 yr
Excavatum
*5 yr
Mild respiratory distress Fever workup
8 yrs
II
1 yr
Mild respiratory distress
2mo
I
19
38
20
40
8 vr 7wk
21
28
Birth
Mild respiratory distress
1 mo
I
22
40
Birth
Mild respiratory distress
II
23
40
6mo
Mild respiratory distress
2 yr 7mo
I
*Incomplete resection done elsewhere at 3 months of age, presented with recurrent respiratory distress and pectus excavatum.
Pathology
Case 3
Lesions were classified (according to Stocker) as follows: 13 with type I (57%), 6 with type II (26%), and 4 with type III (17%) lesions. The right lung was affected twice as frequently as the left, with lesions distributed equally among all lobes. The classification in nonsurvivors were one type I, one type II, and three type III lesions. The two aborted fetuses had type III lesions, one had total bilateral and the other had whole right lung involvement. Overall, only one patient with a type III lesion has survived.
A diagnosis of CAM was made at 30 weeks’ gestation. Other findings included polyhydramnios and ascites. An emergency cesarean section was performed for fetal distress. A hydropic male (Fig 3) was born with Apgars of 2 and 3 at I and 5 minutes, respectively. A left lower lobectomy was immediately performed and a 78-g type II CAM resected. The patient died at 22 hours of life. An autopsy demonstrated associated pulmonary hypoplasia.
Outcome The delay between presentation of symptoms and age at operation was less than 1 week in 7 cases, within 1 year in 8 cases, and more than 1 year in 6 cases. The three patients who died within the first 24 hours postoperatively with documented pulmonary hypoplasia as well as a CAM underwent resection immediately after birth for severe respiratory distress. CASE REPORTS
Case 8 Obstetrical ultrasound showed a fetus having two large thoracic cysts and polyhydramnios at 30 weeks’ gestation, compatible with either CAM or congenital diaphragmatic hernia. Fetal echocardiography was normal and no other anomalies were detected. While fetal intervention was being considered, premature rupture of membranes occured at 34 weeks’ gestation. Repeat ultrasound clearly demonstrated a CAM with two large cysts and mediastinal shift (Figs 4 and 5). Cyst aspiration was attempted to try to lessen medistinal shift, to reduce lung compression, and to relieve polyhydramnios. Because of fetal position and movement, only one cyst could be aspirated and it was not possible to insert “pig-tail” catheters. Spontaneous delivery occurred 2 days later with severe perinatal respiratory distress. At 3 hours of life, a left pneumonectomy was performed. Transient improvement in oxygen saturation from 70% to 90% on 100% oxygen was followed by a rapid deterioration and the patient died at 8 hours of life. The total left lung weight was 38 g with the upper lobe having pulmonary hypoplasia and the lower a type I CAM.
Case 2 An infant had a right thoracic Iesion consistent with a CAM diagnosed at 18 weeks’ gestation. The mass caused mediastinal shift but there was no polyhydramnios, ascites, or other signs of hydrops. Possible regression was taken into account in prenatal counseling. The lesion was followed-up and progressively decreased in size from 4.5 to 2.7 cm (Fig 1) by 30 weeks. After term delivery, there were no respiratory symptoms and the CXR was almost normal (Fig 2). A CXR repeated at 3 months of age showed some mass effect, but the infant continued to be asymptomatic. An elective lobectomy was performed and a 31-g right lower lobe type I CAM was removed.
Case 14 A boy first presented at 3 years of age with an episode of croup. On examination there were decreased breath sounds and an end expiratory wheeze in the left upper chest. A CXR showed a cystic lesion in the left upper lobe that was interpreted as congenital lobar emphysema (Fig 6). Bronchoscopy was consistent with this diagnosis and he was followed-up for 2 years without any symptoms or change in the CXR. However, a CT scan was consistent with a CAM (Fig 7), so a left upper lobectomy was performed which confirmed a type I CAM.
978
NEILSON ET AL
Fig 1. Fetal ultrasound of type I CAM in case 2. On the longitudinal view (A) taken at 18 weeks are arrows overlying the fetal spine and showing the 4.5-cm mass. The transverse view(B) taken at 27 weeks shows the right thoracic mass has decreased to 3.0 cm, but still shifts the heart toward the left thorax.
DISCUSSION
This series was collected over a period during which routine prenatal ultrasound screening became widely available. Compared with most previous reports,3,6,7it is a large series and represents a relatively homogeneous population because it encompasses only a IO-year period. Ultrasound scanning correctly diagnosed CAM in the 10 prenatal cases who either underwent resection or therapeutic abortion. Ultrasound was helpful in recognizing neonates at high risk for perinatal respiratory distress. Prenatal diagnosis permitted transfer to
Fig 2.
CXR of case 2 showing faint cysts in right lower lobe.
a tertiary center for high risk obstetrical and neonatal care. As a result, all symptomatic neonates were operated on promptly. In contrast, those who presented in the postnatal period did not have the benefit of prompt recognition. Approximately 40% of these neonates had respiratory symptoms at birth. However, this delay did not appear to affect morbidity or mortality. The prenatal factors associated with a poor outcome were the detection of polyhydramnios, ascites,
Fig 3. Radiograph of cese 3 showing left lower lobe type II CAM, mediastinal shift, and ascites.
CONGENITAL ADENOMATOID
979
MALFORMATION
Fig 4. Fetal ultrasound of case 8 showing 5.2-cm type I CAM before cyst aspiration at 33 weeks’ gestation.
or mediastinal shift. As well, the more hyperechoic type III CAM, which has a much worse prognosis, could be differentiated from cystic forms of CAM by fetal ultrasound. Ultrasound also was able to detect two fetuses with multiple life-threatening anomalies. These were both associated with type III CAM lesion as well as polyhydramnios and ascites. However, it is important not to overestimate the poor prognosis of a prenatally detected lesion because there were two survivors with polyhydramnios that was only transient. Likewise, in one case, the CAM progressively decreased in size; in another case, associated mediastinal shift disappeared. Partial regression in utero of CAM has also been documented in two case reports.’ The three postnatal deaths in this series were associated with perinatal respiratory arrest with emergency resection undertaken as a last resort. All three
Fig 5. CXR of case 8 at birth showing left lower lobe CAM and mediastinal shift.
Fig 6. Preoperative CXR of case 14 erroneously congenital lobar emphysema.
interpreted
as
died within the first postoperative day, often with a brief postoperative honeymoon period during which respiratory parameters improved. All were later shown to have pulmonary hypoplasia secondary to the CAM. Drainage of large space occupying cysts to reverse fetal hydrops is physiologically sound. Repeat needle aspirations or insertion of in utero thoracoamniotic shunts have been described, occasionally with successful outcome.9-11 Fetal “chest tubes” have also been found to reverse polyhydramnios and hydrops in fetal hydrothorax, which also acts as a space occupying lesion.“,” The management of solid type III CAM in the fetus would be more difficult, requiring open fetal surgical intervention as reported by Harrison et al for diaphragmatic hemia.14 Some cases with less severe
Flg 7. Chest CT scan of case 14 documenting left upper lobe CAM with a large unilocular cyst and several smaller cysts anteriorly and adjacent to the mediastinum.
980
NEILSON ET AL
lung hypoplasia may also benefit from perioperative extracorporal membrane oxygenation support. In contrast to those who died, all infants without a preoperative respiratory arrest have survived. There was one survivor with a type III lesion, which is usually associated with a fatal outcome. Adzick et al4 has reported one other survivor with a type III lesion, In the antenatal diagnosis group, one half of the survivors remained asymptomatic in the newborn period; hence, these were really incidentally diagnosed lesions. However, we would recommend prompt surgery after diagnosis of CAM in all cases. In this series, one infant is awaiting resection of a presumed CAM and another 6-year-old patient has been followed-up from birth with a suspected CAM. Because of the reports of malignancy associated with CAM in young adults,‘S-‘8 as well as the risk of pulmonary
infections, we feel it is important to resect all suspected CAM lesions. In summary, CAh4 has a spectrum of severity, and even when diagnosed in utero it can either regress and be asymptomatic at birth, or can progress to cause either fetal death or neonatal respiratory arrest from pulmonary hypoplasia. Serial ultrasound examinations are required to assess prognosis and allow counseling; however, polyhydramnios and fetal ascites require some form of intervention if death is to be avoided.
ACKNOWLEDGMENT The authors thank Dr Gisele Lussier, radiologist, who provided the fetal ultrasounds in case 2 and was instrumental in the referral for prenatal counseling.
REFERENCES 1. Stocker JT, Madewell JE, Drake RM: Congenital cystic adenomatoid malformation of the lung: Classification and morphologic spectrum. Hum Path01 8:155-171,1977 2. Ch’in KY, Tang MY: Congenital adenomatoid malformation of one lobe of a lung with general anasarca. Arch Path01 48:221229,1949 3. Beluffi G, Brokensha C, Kozlowski K, et al: Congenital cystic adenomatoid malformation of the lung. Presentation of 16 cases. Fortschr Rontgenstr 150:523-530,1989 4. Adzick NS, Harrison MR, Glick PL, et al: Fetal cystic adenomatoid malformation: Prenatal diagnosis and natural history. J Pediatr Surg 20:483-4881985 5. Morin C, Filiatrault D, Russo P: Pulmonary sequestration with histologic changes of cystic adenomatoid malformation. Pediatr Radio1 18:130-l’32,1989 6. Ostor AG, Fortune DW: Congenital cystic adenomatoid malformation of the lung. Am J Clin Path01 70:595-604,1978 7. Wesley JR, Heidelberger KP, DiPietro MA, et al: Diagnosis and management of congenital cystic disease of the lung in children. J Pediatr Surg 21:202-207,1986 8. Saltzman DH, Adzick NS, Benacerraf BR: Fetal cystic adenomatoid malformation of the lung: Apparent improvement in utero. Obstet Gynecol71:1000-1002,1988 9. Chao A, Monoson RF: Neonatal death despite fetal therapy for cystic adenomatoid malformation: A case report. J Reprod Med 35:655-657,199O
10. Nugent CE, Hayashi RH, Rubin J: Prenatal treatment of type I congenital cystic adenomatoid malformation by intrauterine fetal thoracentesis. J Clin Ultrasound 17:675-677,1989 11. Clark SL, Vitale DJ, Minton SD, et al: Successful fetal therapy for cystic adenomatoid malformation associated with second-trimester hydrops. Am J Obstet Gynecol157:294-295,1987 12. Longaker MT, Laberge J-M, Dansereau J, et al: Primary fetal hydrothorax: Natural history and management. J Pediatr Surg 24:573-576,1989 13. Rodeck CH, Fisk NM, Fraser DI, et al: Long-term in utero drainage of fetal hydrothorax. N Engl J Med 319:1135-1138,1988 14. Harrison MR, Adzick NS, Longaker MT, et al: Successful repair in utero of a fetal diaphragmatic hernia after removal of herniated viscera from the let thorax. N Engl J Med 3221582-1584, 1990 15. Sheffield EA, Addis BJ, Corrin B, et al: Epithelial hyperplasia and malignant change in congenital lung cysts. J Clin Path01 40:612-614,1987 16. Prichard MG, Brown PJE, Sterrett GF: Bronchioloalveolar carcinoma arising in longstanding lung cysts. Thorax 39545-549, 1984 17. Weinberg AG, Currarino G, Moore GC, et al: Mesenchymal neoplasia and congenital pulmonary cysts. Pediatr Radio1 9:179182,198O 18. Weinblatt ME, Siegel SE, Isaacs H: Pulmonary blastoma associated with cystic lung disease. Cancer 49:669-671,1982
Discussion A. Hafler (Baltimore, MD): Dr Nielson, I listened to your report with great interest. I kept waiting for you to give us the punchline, which your abstract says is “to find out whether prenatal diagnosis would influence the management of this lesion.” I hope you will indicate to us whether the prenatal studies have altered your management of the patients. I say this because your data show us clearly that one of the type
III patients survived. It also shows us that there can be marked changes even in the early prenatally diagnosed patient in the course of the further development of the lung. I urge caution in supporting abortion because I have had one patient I was quite certain would not survive with bilateral cystic lesions. Because the parents did not wish to interrupt the pregnancy, the patient was carried to term, was
CONGENITAL ADENOMATOID
MALFORMATION
delivered, and is now 8 years old. The patient did require initial resection on one side and, subsequently, has required resection of a small cystic area on the other. It seems to me that we must be very careful in interpreting this. I believe it is important to get a better understanding of the pathophysiology, but I find it difficult to try to use this in a retrospective way. I believe that we should allow these babies to tell us after they’re born whether they are capable of surviving. The concern I have is that the polyhydramnios is not necessarily as you’ve shown us: that is there’s going to be bilateral hypoplasia of the lungs that will not allow survival. I would also like to ask you about hydrops. Do you have a good explanation for that and is it a reversible component also? These studies are very important, but I think that we shouldn’t jump to conclusions until we have a better understanding of the pathophysiology of the lesions. LR. Neilson (response): I’d like to thank Dr Haller for prompting me to stress a very important part of this paper. This concerns the management of CAM once an in utero diagnosis is made. We studied which
981
factors were important in the outcome of CAM, in order to allow more effective prenatal counseling. In this series, there were only two cases that were terminated mainly for multiple associated lifethreatening anomalies, ie, renal agenesis, rather than for the CAM. All of the other fetuses were followed to term. Detection of a type III lesion is not an indication for termination, especially because there was one survivor in this series with a documented type III lesion. Of all prognostic factors, the size of the cyst probably is the most important, because this leads to shift of the mediastinum and indirectly influences the pathogenesis of both polyhydramnios and ascites. We attempted cyst aspiration in one fetus; however, this was probably done too late in pregnancy. There are several case reports in the literature of successful treatment of polyhydramnios, as well as the regression of fetal hydrops. All of the patients in this series who died postoperatively had documented pulmonary hypoplasia. Obviously, if progress is to be made in preventing mortality, fetal intervention will need to be further assessed.