American Association for

Thoracic Surgery

68TH ANNUAL MEETING

APRIL 18-20, 1988

Scientific Program

MONDAY MORNING, April 18, 1988

8:30 a.m. Business Session (Limited to Members)

8:45 a.m. Scientific Session - Ballroom

1. Myotomy for Reflux Induced Cricopharyngeal Dysphagia: Five Year Review

ROBERTO. HENDERSON, WED AD M. HANNA*,

GARY MARRYATT* and ROBERT F. HENDERSON*

Toronto, Ontario, Canada

The cricopharynx is the proximal competent valve preventing reflux to the pharynx and possible aspiration. Despite this 25 patients, all of whom have severe reflux induced cricopharyngeal dysphagia were treated by myotomy and followed for five or more years. Patient selection was a) continued dysphagia following hernia repair (7); b) debility contraindicating hernia repair (12); c) dominant dysphagia without symptomatic reflux (6).

Preoperative evaluation confirmed reflux using a combination of history, manometry, common cavity pH, endoscopy, and radiology. Symptomatical-ly the dysphagia was severe producing weight loss in 42% and respiratory infections in 52%. In the older patient it was severe enough to risk laryngeal obstruction and anoxic death. Past studies have shown that 1.5% of patients have persistence of cricopharyngeal dysphagia following effective hernia repair. All of these had severe preoperative dysphagia.

The investigative findings included radiologic aspiration (80%); reflux 72%; endoscopic incompetence 100%; cricopharyngeal incoordination 91%.

Surgery was under local anesthesia. There was no mortality or morbidity. In follow-up none have reflux induced aspiration pneumonia. Pathologic studies show muscle degeneration compatible with inflammation.

Symptomatically 19 (76%) are asymptomatic; 4 (16%) have minor residual dysphagia; 1 (4%) has moderate dysphagia and 1 is not improved.

In this group failure to respond to medical and surgical management may be due to permanent inflammatory cricopharyngeal pathology. Effective relief of major symptoms has been obtained in 92% of patients with no significant side effects from surgery.

*By Invitation

2. A Comparison of Endoesophageal Devices: Improved Results with the Atkinson Tube

JOHNM. KRATZ*, CAROLYNE. REED*,

FRED A. CRA WFORD, MARTHA R. STROUD* and

EDWARD F. PARKER

Charleston, South Carolina

In the interval January 1, 1973 to December, 1986, we have palliated 40 patients with unresectable carcinoma of the esophagus by placement of Celestin (Group I, n = 12), Proctor-Livingston (Group II, n = 12) or Atkinson (Group III, n = 16) endoesophageal tubes. Age, race, sex, weight loss, location of tumor and duration of symptoms were similar for each group. Laparotomy was required in 100% of Group I patients, 42% of Group II and none of Group III. Postoperative complications (aspiration, sepsis, reflux, and pneumonia) occurred significantly less often in Group III (13%) than in Group I (58%) (p < .05) or II (67%) (p < 0.05). Hospital mortality was lower in Group III (6%) than Group I (42%) (p < .05) or II (42%) (p < .05). Hospital stay was shorter for Group III (4 ± 1 day) than for Group I (16 ± 4) (p < .01) or Group II (14 ± 2) (p < .05). Eighty-one percent of Group III patients reported good swallowing as opposed to 50% in Group I (p = n.s.) and Group II (p < .05). Survival was longer in Group III (108 ± 19 days) than in Group I (48 ± 9) (p < .05) or II (39 ± 8) (p < .01). Complications following discharge were fewer in Group III (33%) than I (71%) or II (43%) but the difference was not significant.

When patients were grouped into those requiring or not requiring a laparotomy for tube placement, then hospital complications (p < .01), operative mortality (p = n.s.), hospital days (p < .01) and home complications (p = .06) all occurred more frequently in the laparotomy group. In addition, patients whose tube did not cross the GE junction had less postoperative complications including aspiration (p = n.s.), sepsis (p < .05), reflux (p < .05) and pneumonia (p = n.s.). This data suggests that the Atkinson tube provides a superior method of palliation in patients with unresectable carcinoma of the esophagus and that its lower rate of complications may at least in part be due to the lack of need for a laparotomy for tube placement and the frequent lack of need for the tube to cross the GE junction.

*By Invitation

3. Does Open Lung Biopsy Affect Treatment in Patients with Diffuse Infiltrates?

WILLIAM A. WALKER*, F. HAMMOND COLE, JR.*,

ALIM KHANDEKAR*, SAADE S. MAHFOOD*

and DONALD C. WATSON

Memphis, Tennessee

The decision to perform open lung biopsy in the evaluation of a diffuse pulmonary infiltrate is based on the probability that this examination will yield specific information leading to a change in treatment. The role of this procedure remains controversial and many clinicians are reluctant to allow this invasive procedure without assurances that results will lead to a change in therapy for a significant number. To evaluate the impact of open lung biopsy on diagnosis and treatment of diffuse pulmonary infiltrates, a retrospective review of 50 patients undergoing this procedure at three university-affiliated hospitals during a recent six year period was performed. There were 31 males and 19 females, average age 55 years. Biopsy yielded a specific diagnosis in 18 (36%) patients, and a change in therapy in 28 (56%) patients. A complication developed in 9 (18%) patients; 5 (10%) were directly related to the biopsy procedure. Seven patients expired. Immune status in 19 (38%) patients was compromised. A specific diagnosis was obtained in 11 (58%) immunocompromised patients and a change in therapy occurred in 15 (79%) patients after biopsy. A specific diagnosis was obtained in only 7 (23%) of the 31 non-compromised patients and therapy was changed in 14 (45%) patients in this group (p < 0.02 compromised vs. non-compromised). Morbidity and mortality were not significantly different between the two groups. A non-specific diagnosis led to a change in therapy as frequently as a specific diagnosis in both compromised and non-compromised groups. Open lung biopsy in the patient with a diffuse pulmonary infiltrate is an accurate diagnostic tool and frequently leads to a change in patient treatment. The procedure can be performed with acceptable morbidity and mortality in immunocompromised and non-compromised patients.

*By Invitation

4. Trans-Tracheal Aspirate and Fine Needle Aspiration Biopsy for the Diagnosis of Pulmonary Infection in Immuno-compromised Patients

FABRIZIO de VIVO*, GERALD D. POND*,

BIRGER RHENMAN*, TIMOTHY B. ICENOGLE*,

M. ANDRE VASU* and JACK G. COPELAND

Tucson, Arizona

Between 1979 and 1987 a total of 65 patients with heart (61) or heart/lung transplants (4) underwent trans-tracheal aspiration (TTA) and/or fine needle aspiration (FNA) to isolate the causative pathogen of suspected pulmonary infection. Eighty-two TTA's and 47 FNA's were performed. Twenty-three patients underwent both TTA and FNA sequentially. Table I shows the results of TTA alone, FNA alone, and combined TTA/FNA. Table II shows the sensitivity, specificity, and accuracy of TTA, FNA, and combined TTA/FNA.

TABLE I
	True		False		Total
	+	-	+	-	Procedures
TTA	39	25	1	14	82
FNA	34	9	-	4	47
TTA&FNA	18	3	-	2	23
TABLE II
	Sens		Spec			Ace (%)
TTA	70%		96%			78%
FNA	89%		100%			91%
TTA&FNA	90%		100%			91%

TTA was the first examination performed in all patients regardless of the chest radiographic findings. This accounts for the lower sensitivity of TTA (70%) compared to FNA (89%). FNA requires a fluoroscopically identifiable lesion which can be used as a "target".

Moderate bleeding at the cricothyrotomy site was the only complication attributable to TTA (1%). A total of 11 complications resulted from FNA (23%), including 10 pneumothoraces, one of whom had associated air embolism.

The commonest pathogens isolated were Nocardia (34%), Legionella (18%), and Coccidiodomycosis (8%).

Because TTA is technically simple and associated with low morbidity, the authors conclude that for immunocompromised patients with suspected pulmonary infection it should be performed first in spite of the procedure's low sensitivity. If TTA fails to document a pathogen, then FNA can be done. FNA has a higher sensitivity but is associated with a greater morbidity.

10:05 a.m. Intermission - Visit Exhibits

*By Invitation

10:50 a.m. Scientific Session - Ballroom

5. A Clinical Study on Cerebral Circulation During Extracorporeal Circulation

YASUHIRO SOMA*, TAKASHI HIROTANI*,

RYOHEI YOZU*, TAKAHIKO MISUMI*,

KOZO KAWADA* and TADASHI INOUE*

Tokyo Japan

Sponsored by: HITOSHI MOHRI,

Tokyo, Japan

The goal of this study is to clarify the relationship of cerebral blood flow (CBF) to extracorporeal circulation (ECC) flow and mean arterial pressure (MAP) during nonpulsatile ECC under moderate hypothermia (27-28 C) to know the lower limit of ECC flow and MAP to meet the oxygen demand of the brain. The CBF was determined by Argon saturation and desaturation method after Pevsner et al using Mass-spectrometer in 21 adult patients undergoing cardiac operation. The CBF during ECC was such: 1) 33.8 + 8.9 ml/100g brain/min on average and ECC flow passive, that is, proportional to ECC flow (r = 0.73, p<0/01) while it was between 40 to 70 ml/Kg/min, 2) independent to MAP; showed no significant change while MAP ranged from 30 to 94 mmHg, that is, the lower limit of the cerebral autoregulation extended down to 30 mmHg. Cerebral oxygen consumption did not demonstrate significant decrease as far as CBF of 25 ml/100g brain/min or ECC flow of 40 ml/Kg/min was maintained. All of the patients survived the operation and have no postoperative cerebral damage. In conclusion, CBF was ECC flow passive and was maintained by the cerebral autoregulation down to MAP of 30 mmHg. We believe this is the first report that clearly demonstrated CBF to be ECC flow passive by measuring CBF itself.

*By Invitation

6. Extracorporeal Membrane Oxygenation for the Treatment of Neonatal Respiratory Failure: A Report of 50 Cases

MICHAEL G. MORONT*, NEVINM. KATZ*,

MAR TIN KESZLER*, MARC S. VISNER *,

GREGORYR. HOY*, JOHN J. O'CONNELL*,

CYNTHIA COX* and ROBERT B. WALLACE

Washington, DC

Despite progress in neonatal intensive care, pulmonary failure remains an important cause of morbidity and mortality in high risk newborns. In the past few years Extracorporeal Membrane Oxygenation (ECMO) has become a therapeutic alternative for term neonates with life threatening respiratory failure. From February 1985 through June 1987, 50 newborn infants who failed maximal conventional ventilatory therapy (80% predicted mortality) were treated with ECMO according to the following entry criteria: gestational age < 36 wks., birth weight > 2 kg, p0₂ < 50 torr (AaD0₂ > 630 torr) for 2 hours or p0₂ < 60 torr (AaD0₂ > 620 torr) for 8 hours. Patient characteristics included: mean birth weight 3.28 ± 0.56 kg, mean gestational age 39.6 ± 1.7 weeks, mean 1 min. Apgar score 4.5 ± 7.2, and mean 5 min Apgar score 6.2 ± 2.5. Diagnoses for which ECMO was instituted were meconium aspiration (62%), primary pulmonary hypertension (8%), respiratory distress syndrome (8%), blood aspiration syndrome (8%), congenital diaphragmatic hernia (6%), congenital pneumonia (6%), and pulmonary hypoplasia (2%). To prevent pulmonary atelectasis, lung management during ECMO consisted of high Positive End Expiratory Pressure (PEEP), (8-14 cm H₂O) in 94% of patients. Arterial blood gases and ventilator settings before, during, and after ECMO are presented (mean with SD) (* = p < 0.05 for difference from pre ECMO parameter).

					PEEP
	P02(torr)	PH	PC02(torr)	PIP(cmH.O)	(cmH20)	FiO2	IMV
PreECMO	34.5±14.5	7.38±0.21	41.5±19.6	46.8±9.9	4.6 ±1.6	1.0±0.00	101.2±22.1
1 hr on ECMO	*180.4±107.4	*7.52±0.07	*26.5±5.5	*22.8±1.6	*10.3±2.5	*0.21±0.02	*11.8±2.9
1 hr after ECMO	*78.4±22.1	7.39±0.10	37.4±10.7	*25.2±3.9	5.6±1.2	*0.42±0.17	*41.3±12.6

PIP-Peak Inspiratory Pressure

FiO₂-Fraction Inspired Oxygen

IMV-lntermittent Mandatory Ventilation/min.

Overall long term patient survival was 90%. Mean PEEP maintained during ECMO was significantly greater than that reported by the National ECMO Registry (N.E.R.) (10.3 vs. 4.5 cmH₂0, p < 0.01). Mean pre ECMO p0₂ was significantly lower than that reported from the N.E.R. (34.5 vs. 42.6 torr, p < 0.01) despite slightly higher levels of ventilator support (PIP 46.8 vs. 44.2 cmH₂0, N.S.). Mean duration of ECMO support was significantly shorter compared to the N.E.R. (84.3 vs. 117.2 hrs., p < 0.01). In contrast to reports in the literature only 27% of our patients demonstrated an important increase in lung opacification while on ECMO. Mean time to extubation, length of hospital stay, and incidence of complications compared favorably to N.E.R. data.

We conclude that ECMO is a remarkably effective modality to reverse severe neonatal respiratory failure and that excellent survival can be achieved, despite the critical condition of these patients. The use of high PEEP during neonatal ECMO minimizes pulmonary atelectasis and may importantly decrease the necessary duration of ECMO support.

11:30 a.m. Presidential Address

Paul A. Ebert, M.D., Chicago, Illinois

12:15 p.m. Adjourn for Lunch - Visit Exhibits