Journal of Applied Physiology Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 83: 1733-1740, 1997;
8750-7587/97 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Imanaka, H.
Right arrow Articles by Kacmarek, R. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Imanaka, H.
Right arrow Articles by Kacmarek, R. M.

Vol. 83, Issue 5, 1733-1740, 1997

Recovery of diaphragmatic function in awake sheep after two approaches to thoracic surgery

Hideaki Imanaka, William R. Kimball, John C. Wain, Masaji Nishimura, Kenichi Okubo, Dean Hess, and Robert M. Kacmarek

Departments of Anesthesia and Critical Care, Respiratory Care, and Surgery, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 02114

ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
ACKNOWLEDGEMENTS
FOOTNOTES
REFERENCES

ABSTRACT

Imanaka, Hideaki, William R. Kimball, John C. Wain, Masaji Nishimura, Kenichi Okubo, Dean Hess, and Robert M. Kacmarek. Recovery of diaphragmatic function in awake sheep after two approaches to thoracic surgery. J. Appl. Physiol. 83(5): 1733-1740, 1997.---Video-assisted thoracoscopic surgery (VATS) is replacing thoracotomy, but no study has addressed the extent or duration of VATS-induced diaphragmatic alteration. We hypothesized that VATS would impair diaphragmatic function less and return diaphragmatic function faster than thoracotomy. In eight sheep, sonomicrometers were randomly implanted on the right costal diaphragm via VATS or thoracotomy. Diaphragmatic resting length, shortening fraction, and respiratory function were measured weekly during quiet breathing (QB) and CO2 rebreathing for 4 wk. For VATS, shortening fraction was smallest on postoperative days 1 (POD 1) (6.4 ± 3.4 and 12.9 ± 8.7% during QB and 10% CO2 rebreathing, respectively) and 7 (6.3 ± 3.4 and 16.9 ± 4.0% during QB and 10% CO2 rebreathing, respectively) and recovered by 3 wk (13.2 ± 1.8 and 28.9 ± 8.0% during QB and 10% CO2 rebreathing, respectively). For thoracotomy, shortening fraction at 10% CO2 rebreathing was smaller on PODs 1, 7, 14 (15.9 ± 7.1, 13.6 ± 5.4, and 19.0 ± 6.9%) than on POD 28 (29.9 ± 8.2%), but not during QB on POD 1 or 7 (7.5 ± 3.8 and 3.4 ± 2.6%) compared with POD 28 (10.7 ± 8.7%). Shortening fraction did not differ between surgeries. There was no group difference in minute ventilation, respiratory rate, transdiaphragmatic pressure, or esophageal and gastric pressures. In conclusion, although shortening fraction recovered faster for VATS, this translated into insignificant functional differences.

sonomicrometers; video-assisted thoracoscopic surgery

INTRODUCTION

DIAPHRAGMATIC FUNCTION is depressed by thoracotomy (9, 10, 18, 21, 22, 28) or laparotomy (3, 4, 24, 25). Recent improvements in optics have allowed many operations previously requiring these surgical approaches to be performed using endoscopic techniques. Video-assisted thoracoscopic surgery (VATS) has been reported to demonstrate improvements over classic open thoracotomy, leading to reduced analgesic requirements (11, 16, 30), shorter hospital stays (2, 7, 16), fewer complications (2, 7, 15, 16), and smaller reductions of vital capacity or forced expiratory volume in 1 s (16, 30). However, other studies report no difference in forced vital capacity or forced expiratory volume in 1 s between VATS and muscle-sparing thoracotomy (11) or in clinical outcome for patients randomized to receive lobectomy by VATS or thoracotomy (15). Although benefits after VATS may relate to factors inherent in its approach, differences in patient groups, incision length, amount of tissue excised, or underlying disease could be responsible for these reported advantages.

Although VATS appears to provide benefits over thoracotomy, no study has addressed the extent or duration of diaphragmatic functional alterations after VATS, especially where such previously described variables have been strictly controlled, possibly because diaphragmatic function is difficult to address indirectly (23, 28). We previously evaluated recovery of diaphragmatic function after thoracotomy and implantation of sonomicrometers onto the sheep diaphragm and found that diaphragmatic contraction was depressed substantially for 14 days, then recovered fully after 28 days (28). Employing this approach, we compared the recovery of diaphragmatic function and ventilatory parameters in sheep during quiet breathing (QB) and CO2 rebreathing using thoracotomy or VATS to implant sonomicrometers onto the costal diaphragm. This approach permitted strict control of the surgical and anesthetic interventions, pain relief, animal study groups, and measurement interventions while eliminating factors inherent in clinical studies.

On the basis of clinicians' almost uniform opinion that VATS is superior to open thoracotomy and the available literature (2, 7, 11, 15, 16, 30), we postulated that recovery of diaphragmatic function after VATS would be substantially faster, i.e., complete by 14 days. In our previous study of diaphragmatic recovery after thoracotomy, we noted that diaphragmatic shortening during QB improved by 5% of diaphragmatic resting length (length at functional residual capacity, LFRC; from 8.0 to 13.0%) between postoperative days 14 (POD 14) and 28, with a standard deviation of ~1.25% LFRC (28). Employing these data, we predicted that four animals per group could reveal a difference between VATS and thoracotomy, even if the error were nearly twice as large, similar to standard deviations of our previous studies (9, 21, 22, 27). An analysis at 10% end-tidal CO2 (ETCO2) supported this prediction. We reasoned that, if recovery of diaphragmatic function after VATS demonstrated smaller differences, it was unlikely to be the primary factor responsible for the clinical impression that recovery after VATS is superior to that after thoracotomy.

MATERIALS AND METHODS

Surgical procedures. This study was approved by the Animal Care Committee of Massachusetts General Hospital. The experimental methods have been described previously (9, 10, 21, 22, 27, 28). Briefly, eight Hampshire sheep (28-35 kg) underwent a tracheostomy during general anesthesia (1.5-2% halothane). After exposure of the left third parasternal muscle, two triangular 10-mm Dacron patches (10), each containing a sonomicrometer crystal and electromyogram (EMG) electrode, were sutured 8 mm apart onto the muscle after the sonomicrometer and EMG electrode had been embedded into the muscle. Wires were tunneled subcutaneously, then exteriorized through separate skin incisions near the posterior thoracic spine. Local anesthesia (10 ml of 2% lidocaine) was infiltrated around the incisions. Five days later, during halothane anesthesia, sonomicrometer crystals and EMG leads were implanted onto the right hemidiaphragm. This date was counted as POD 0. Sheep were randomly assigned to VATS or thoracotomy just before their arrival in the operating room. Interventions such as anesthesia, airway management, and postoperative care were identical for both surgical techniques. A bronchial blocker (10-Fr Fogarty embolectomy catheter) was passed alongside the tracheostomy tube, then advanced with fiber-optic bronchoscopic guidance and inflated (~5 ml air) to occlude the right main stem bronchus. Unilateral ventilation was commenced at the start of the operation to facilitate surgical exposure.

In the VATS group (n = 4), four 1.5-cm skin incisions were made: one in the right third to fourth interspace for the videoscope and three in the right seventh and eighth interspaces for surgical instruments. The procedure required 1.0-1.5 h. In the thoracotomy group (n = 4), 12-15 cm of the skin and muscle of the ninth thoracic interspace were incised. The lower border of the intercostal muscle including the periosteum was lifted from the upper border of the 10th rib, and the pleura was opened. The rib was separated gently with a rib retractor. This procedure required ~1 h.

Sonomicrometers were implanted on the dome of the right diaphragm between costal muscle fibers within 2-3 cm of the central tendon for both surgical groups. Each diaphragmatic crystal and EMG electrode was attached to a triangular 20-mm Dacron patch. Transducers were embedded in the muscle 12-20 mm apart along the direction of the muscle fibers, then patch corners were sutured to the diaphragm with 4-0 silk sutures. Satisfactory functional alignment of the crystals was confirmed during surgery by measuring the sonomicrometer signal and then by direct tetanic stimulation of the diaphragm. Cables for both surgical groups were brought through the chest wall at the lateral costophrenic junction, then tunneled to the posterior paravertebral region, where they were exteriorized through separate skin incisions. After the thorax was closed, the pleural space was evacuated via a chest tube during positive-pressure lung expansion. Lidocaine (20 ml, 2%) was infiltrated subcutaneously around the incisions. Sheep were given antibiotics (Cefazolin, 1 g) for 4 days. No additional analgesics were necessary.

Measurements. Sheep were fasted for 24 h, then studied on preoperative day 1 (POD -1), POD 1, and every 7 days for 4 wk, as previously described (9, 21, 22, 28). Diaphragmatic measurements commenced on POD 1. Measurements were recorded during QB and CO2 rebreathing. Sheep stood unsedated and breathed through a cuffed tracheostomy tube (7 mm ID, Portex, Keene, NH) via a heat-moisture exchanger (Thermovent 600, Portex) connected to a Fleisch pneumotachograph (model 3700A, Hans Rudolph, Kansas City, MO) and a pressure transducer (model MP-45, ±2 cmH2O, Validyne, Northridge, CA). Tidal volume (VT) was obtained by integration of the airflow signal (model 8815A, Hewlett-Packard, Waltham, MA) and calibrated with a 1-liter syringe.

After topical anesthesia (10 ml of 2% lidocaine), two balloon-tipped catheters (model 85842, National Catheter, Mallinckrodt, Argyle, NY) were inserted transnasally into the stomach and esophagus. Each catheter was connected to a pressure transducer (MP-45, ±100 cmH2O, Validyne), calibrated at 20 cmH2O using a water manometer. The esophageal balloon contained 1 ml of air and the gastric balloon 2 ml of air. The esophageal balloon position was adjusted to minimize differences between airway and esophageal pressure (Pes) during airway occlusions (1). Instantaneous differences between Pes and gastric pressure (Pga) defined transdiaphragmatic pressure (Pdi). Pga was measured during inspiration (Pgain) and expiration (Pgaex).

Muscle shortening was measured via a sonomicrometer (model 120, Triton Technology, San Diego, CA). Signals from the costal and parasternal EMG electrodes were preamplified and band-pass filtered (30 Hz-3 kHz, model P511K, Grass Medical Instruments, Quincy, MA). These signals were rectified using a Paynter filter with a 100-ms time constant. EMG systems were calibrated with a 0.2-mV sine wave of 80 Hz (model 182, Wavetek, San Diego, CA). Expired CO2 concentration was measured with an infrared capnometer (model 2200, Traverse Medical Monitors, Saline, MI). The capnometer was calibrated with 5% CO2-95% O2.

All measurements were recorded on an eight-channel recorder (model WR3600, Graphtec, Irvine, CA) and on a microcomputer at 83 Hz/channel utilizing Windaq software (Dataq Instruments, Akron, OH). Data were analyzed utilizing Windaq playback software.

Experimental protocol. After a control period of QB, sheep were connected to an anesthesia bag containing 4-5 liters of 5% CO2-95% O2. Rebreathing was continued until ETCO2 reached 10%. Sheep recovered for at least 20 min. Three CO2-rebreathing sequences were analyzed during QB and at 8, 9, and 10% ETCO2. Awake animals sometimes breathe irregularly; these irregularities decreased as CO2 level increased. Measurement of five breaths at a specific ETCO2 (usually the lower levels) was not always possible during irregularities. The two runs providing the most acceptable number of breaths at each CO2 level were used for analysis.

Phrenic nerve stimulation (supramaximal voltage, 100 Hz, 200 s, duty cycle 2%, 3-5 s; model S-88, Grass Medical Instruments) via the internal jugular vein at the level of the superior vena cava was performed on PODs 1, 14, and 28. Stimulation was considered optimal when maximal diaphragmatic shortening was achieved with a closed airway. After measurements on POD 28, euthanasia was performed. The location and status of sonomicrometers were examined.

Data analysis and statistical methods. Employing sonomicrometers, we measured costal and parasternal intercostal muscle lengths at end expiration (LFRC) and expressed their shortening as the percent change from the resting length. By using the airflow signal, we calculated respiratory rate (RR), VT, minute ventilation (VE), inspiratory time (TI), and TI-to-total respiratory cycle time ratio (TI/TT). The integrated EMG signal was corrected for TI. All values are the means of 10 breaths, 5 sequential breaths from the 2 best runs. Values are means ± SD. Mean values for each variable were tested by two-way analysis of variance with repeated measures of two factors (surgery and time) across six periods (PODs -1, 1, 7, 14, 21, and 28) during QB and for 10% ETCO2. Multiple comparison testing of means for each individual period was performed using paired Student's t-tests with Bonferroni's correction. Significance was set at P < 0.05.

RESULTS

Eight sheep without operative, postoperative, or autopsy complications were studied. Five additional animals underwent implantation (3 for VATS and 2 for thoracotomy) and survived the 28-day protocol but were excluded from analysis: for VATS, one developed severe fibrotic adhesions to diaphragm crystals, resulting in impaired diaphragm shortening; one showed extensive diaphragmatic muscular fibrosis around one crystal; and one developed a granuloma extending from diaphragm to subcutaneous tissues; for thoracotomy, one developed severe fibrotic adhesions to diaphragm crystals and one showed extensive diaphragmatic muscular fibrosis.

Costal diaphragmatic shortening fraction and segmental length. Representative tracings during QB and 10% ETCO2 are shown in Fig. 1. Data for each surgery are presented in Table 1 for QB and 10% ETCO2 among different postoperative days. For VATS the shortening fractions at QB and 10% ETCO2 on PODs 1 and 7 were smaller than those on POD 28 (P < 0.05). The shortening fraction on POD 21 was almost identical to that on POD 28. For thoracotomy the shortening fraction at 10% ETCO2 was smaller on PODs 1, 7, and 14 than on POD 28 (P < 0.05), but there was no statistical difference in shortening fraction during QB. There was no significant difference in resting length among postoperative days regardless of surgery group, although resting length was significantly longer in VATS than in thoracotomy on all postoperative days (P < 0.01).
Fig. 1. Representative tracings during quiet breathing (A) and at 10% end-tidal CO2 concentration (ETCO2) during CO2 rebreathing (B) after thoracotomy. Top to bottom: segmental length (Lcos and Lps) and EMG (Ecos and Eps) for costal diaphragmatic and parasternal intercostal muscles, esophageal pressure (Pes), gastric pressure (Pga), tidal volume (VT), and exhaled CO2 concentration (CO2).
[View Larger Version of this Image (26K GIF file)]

Table  1.   Ventilatory parameters during quiet breathing and 10% CO2 rebreathing
Preop Postoperative Day
1 7 14 21 28
VATS
 VE, l/min
  QB 6.2 ± 2.2  4.7 ± 0.3  5.2 ± 2.1  7.0 ± 2.9  6.6 ± 3.3  7.8 ± 5.1 
  10% CO2 29.8 ± 2.2  25.1 ± 2.3* 28.8 ± 6.5  29.3 ± 3.3  31.7 ± 4.3  34.6 ± 4.5 
VT, ml
  QB 258 ± 32  235 ± 40  225 ± 42  262 ± 27dagger 241 ± 13  234 ± 53 
  10% CO2 588 ± 121  511 ± 119  613 ± 104  608 ± 67  585 ± 84  587 ± 99 
RR, min-1
  QB 23.9 ± 6.7  20.1 ± 2.1  23.2 ± 9.6  26.3 ± 9.2  27.7 ± 15.4  38.6 ± 34.9 
  10% CO2 52.3 ± 9.9  50.6 ± 7.7  47.1 ± 5.7  48.8 ± 7.5  54.9 ± 9.8  60.7 ± 15.2 
TI/TT
  QB 0.32 ± 0.06  0.34 ± 0.06  0.36 ± 0.11  0.37 ± 0.05  0.35 ± 0.08  0.36 ± 0.11 
  10% CO2 0.45 ± 0.04  0.44 ± 0.04  0.44 ± 0.03  0.45 ± 0.03  0.46 ± 0.02  0.44 ± 0.04 
Pdi, cmH2O
  QB 9.3 ± 3.0  7.4 ± 1.0  6.8 ± 1.9  8.9 ± 2.4  8.3 ± 1.8  7.4 ± 1.4 
  10% CO2 44.3 ± 2.6  39.1 ± 5.3  39.9 ± 7.4  44.9 ± 11.2  47.7 ± 12.6  50.6 ± 6.9 
Pes, cmH2O
  QB 8.2 ± 2.5  6.5 ± 0.8  5.9 ± 1.6  7.9 ± 2.4  7.4 ± 1.9  7.5 ± 3.1 
  10% CO2 52.2 ± 3.7  43.5 ± 5.7  47.4 ± 9.4  50.6 ± 10.6  57.9 ± 19.2  59.4 ± 6.9 
Pgain, cmH2O
  QB 1.4 ± 0.5  1.3 ± 0.3  1.2 ± 0.3  1.3 ± 0.4  1.2 ± 0.1  1.1 ± 0.4 
  10% CO2 0.7 ± 0.3  0.3 ± 0.1  0.6 ± 0.3  0.6 ± 0.1  0.7 ± 0.2  0.5 ± 0.1 
Pgaex, cmH2O
  QB 0.2 ± 0.1  0.3 ± 0.1  0.3 ± 0.1  0.3 ± 0.1  0.4 ± 0.2  0.2 ± 0.1 
  10% CO2 9.8 ± 3.6  6.2 ± 1.2* 9.7 ± 3.5  8.5 ± 1.3  11.9 ± 5.0  10.6 ± 1.3 
Lcos, mm
  QB 20.4 ± 0.7dagger 21.9 ± 2.5dagger 22.4 ± 2.3dagger 22.8 ± 2.7dagger 21.8 ± 1.6dagger
Costal shortening, mm
  QB 1.30 ± 0.71  1.43 ± 0.93  2.05 ± 0.89  3.02 ± 0.67  3.02 ± 0.69 
  10% CO2 2.43 ± 1.58  3.53 ± 0.96  5.02 ± 1.79  6.13 ± 1.69  6.05 ± 1.53 
SFcos, %LFRC
  QB 6.4 ± 3.4* 6.3 ± 3.4* 9.1 ± 3.4  13.2 ± 1.8  13.8 ± 2.9 
  10% CO2 12.9 ± 8.7* 16.9 ± 4.0* 22.9 ± 6.5  28.9 ± 8.0  29.3 ± 6.5 
Lps, mm
  QB 7.7 ± 0.8  8.2 ± 0.7 
SFps, %LFRC
  QB 0.86 ± 1.69  0.22 ± 0.89 
  10% CO2 2.21 ± 6.12  0.67 ± 3.32 
ETCO2
  QB 5.68 ± 0.36  5.43 ± 0.38  5.70 ± 0.39  5.73 ± 0.10  5.55 ± 0.38  5.70 ± 0.18 
Thoracotomy
 VE, l/min
  QB 4.6 ± 0.4  5.3 ± 0.4  4.7 ± 0.9  4.6 ± 0.8  5.0 ± 0.8  6.3 ± 2.7 
  10% CO2 37.9 ± 3.3  28.5 ± 6.3  25.7 ± 3.6  37.2 ± 9.3  34.4 ± 4.4  31.7 ± 3.4 
VT, ml
  QB 203 ± 13  215 ± 51  182 ± 20  203 ± 19  194 ± 36  201 ± 55 
  10% CO2 766 ± 152  611 ± 225  499 ± 76* 709 ± 195  651 ± 143  627 ± 129 
RR, breaths/min
  QB 22.9 ± 1.4  25.2 ± 3.8  26.4 ± 8.1  23.1 ± 3.9  26.2 ± 6.8  34.4 ± 21.6 
  10% CO2 50.5 ± 7.8  48.7 ± 8.1  53.1 ± 13.0  53.5 ± 10.1  54.8 ± 12.4  52.0 ± 10.5 
TI/TT
  QB 0.31 ± 0.10  0.35 ± 0.02  0.32 ± 0.05  0.28 ± 0.07  0.27 ± 0.01  0.32 ± 0.04 
  10% CO2 0.46 ± 0.02  0.45 ± 0.04  0.45 ± 0.03  0.46 ± 0.05  0.47 ± 0.03  0.45 ± 0.06 
Pdi, cmH2O
  QB 6.4 ± 2.4  7.1 ± 2.0  7.5 ± 1.2  7.4 ± 1.3  6.4 ± 0.9  8.1 ± 2.0 
  10% CO2 41.0 ± 5.2  34.6 ± 7.5* 38.9 ± 6.1  46.9 ± 6.1  40.3 ± 1.5  48.2 ± 4.1 
Pes, cmH2O
  QB 5.7 ± 2.2  6.0 ± 1.7  6.6 ± 1.1  6.7 ± 1.6  5.6 ± 0.8  7.1 ± 2.1 
  10% CO2 50.1 ± 5.9  41.9 ± 7.6* 48.4 ± 10.2  58.0 ± 7.3  49.3 ± 3.6  59.8 ± 5.2 
Pgin, cmH2O
  QB 1.1 ± 0.2  1.3 ± 0.6  1.0 ± 0.2  1.0 ± 0.2  1.0 ± 0.3  1.2 ± 0.3 
  10% CO2 0.7 ± 0.3  0.5 ± 0.2  0.7 ± 0.4  0.6 ± 0.3  0.5 ± 0.3  0.6 ± 0.3 
Pgaex, cmH2O
  QB 0.1 ± 0.1  0.2 ± 0.3  0.1 ± 0.1  0.2 ± 0.1  0.3 ± 0.2  0.3 ± 0.2 
  10% CO2 10.8 ± 5.2  9.9 ± 3.5  11.2 ± 4.8  13.1 ± 4.9  11.3 ± 3.9  13.5 ± 5.0 
Lcos, mm
  QB 12.6 ± 2.5  13.5 ± 2.3  12.3 ± 1.3  11.7 ± 2.0  12.9 ± 2.1
  QB 0.92 ± 0.42  0.42 ± 0.26  0.64 ± 0.64  1.04 ± 0.53  1.26 ± 0.94 
  10% CO2 1.82 ± 0.85  1.61 ± 0.31  2.09 ± 0.48  2.61 ± 0.36  3.40 ± 0.70 
SFcos, %LFRC
  QB 7.5 ± 3.8  3.4 ± 2.6  5.6 ± 6.0  9.6 ± 5.9  10.7 ± 8.7 
  10% CO2 15.9 ± 7.1* 13.6 ± 5.4* 19.0 ± 6.9* 25.7 ± 8.5  29.9 ± 8.2 
Lps, mm
  QB 10.3 ± 2.0  10.5 ± 1.6  11.0 ± 1.8 
SFps, %LFRC
  QB 1.62 ± 0.56  1.47 ± 1.08  1.33 ± 0.79 
  10% CO2 8.77 ± 3.22  7.10 ± 2.62  6.31 ± 2.55 
ETCO2
  QB 5.80 ± 0.42  5.43 ± 0.26  5.90 ± 0.29  6.00 ± 0.32  5.83 ± 0.17  5.65 ± 0.26
Values are means ± SD. VATS, video-assisted thoracoscopic surgery; QB, quiet breathing; ETCO2, end-tidal CO2; VE, minute ventilation; VT, tidal volume; RR, respiratory rate; TI/TT, inspiratory time-to-total respiratory cycle time ratio; Pdi, transdiaphragmatic pressure; Pes, esophageal pressure; Pgain and Pgaex, gastric pressure changes during inspiration and expiration; Lcos and Lps, costal and parasternal resting length; SFcos and SFps, costal and parasternal fractional shortening; LFRC, length at functional residual capacity. * P < 0.05 vs. postoperative day 28; dagger P < 0.05 vs. thoracotomy.

Diaphragmatic shortening fraction as a function of VT for VATS and thoracotomy during the recovery period for QB and CO2 rebreathing is shown in Fig. 2. For VATS the shortening fraction at each ETCO2 or VT was smallest on POD 1 and recovered gradually by POD 21. By contrast, the thoracotomy group shortening fraction was smallest on PODs 7 and 14 and then recovered. However, the shortening fraction was not different statistically between VATS and thoracotomy (P = 0.428 and 0.728 for QB and 10% CO2, respectively). The maximal shortening on POD 28 at 10% ETCO2 was identical between groups (29.3 ± 6.5 and 29.9 ± 8.2%). No important difference was found for shortening vs. Pdi (Fig. 3).
Fig. 2. Costal shortening fraction (SFcos) vs. tidal volume during quiet breathing through CO2 rebreathing (8, 9, and 10% ETCO2) on days 1, 7, 14, 21, and 28 after video-assisted thoracoscopic surgery (A) and thoracotomy (B). Numbers at top of curves represent number of days after surgery. Values at 10% ETCO2 are means ± SE.
[View Larger Version of this Image (15K GIF file)]

Fig. 3. SFcos vs. transdiaphragmatic pressure (Pdi) changes on days 1, 7, 14, 21, and 28 after video-assisted thoracoscopic surgery (A) and thoracotomy (B). Numbers at top of curves represent number of days after surgery. Values at 10% ETCO2 are means ± SE.
[View Larger Version of this Image (16K GIF file)]

Breathing pattern and ventilatory parameters. There was no significant difference between VATS and thoracotomy in VE, VT, RR, TI/TT, Pdi, Pes, Pgain, and Pgaex (P = 0.257-0.970, 2-way analysis of variance), except for resting VT on POD 14 (P < 0.05; Table 1). At 10% ETCO2, VE for VATS was smaller on POD 1 than on POD 28. There was no difference in VE among postoperative days for thoracotomy. RR and TI/TT did not change with recovery day (P = 0.108 and 0.840, respectively) or among the surgeries (P = 0.981 and 0.840, respectively). There was no difference between the surgery groups in Pdi and Pes during QB (P = 0.430 and 0.420, respectively). For thoracotomy and 10% ETCO2, Pdi and Pes were smaller on POD 1 than on POD 28, whereas there was no difference during QB.

Parasternal shortening and EMG. Shortening fraction of the parasternal muscle is presented when n = 4 (preoperation and PODs 1 and 7), because the units became dysfunctional later in recovery. The maximal shortening at 10% ETCO2 decreased from preoperation to POD 1 for VATS, whereas values were similar preoperatively and on PODs 1 and 7 for thoracotomy. EMG was not compared between surgeries, because only three sheep maintained good costal EMG throughout recovery and five sheep maintained good parasternal EMG.

Phrenic nerve stimulation and sonomicrometer location. Resting length and maximal shortening during supramaximal stimulation did not change on PODs 1, 14, and 28 (22.6 ± 2.4, 22.4 ± 3.1, and 24.5 ± 1.9 mm and 47.9 ± 5.1, 45.8 ± 10.4, and 50.3 ± 3.2% LFRC, respectively) for VATS, whereas for thoracotomy the maximal shortening (41.9 ± 5.7, 36.0 ± 10.9, and 51.2 ± 14.2% LFRC on PODs 1, 14, and 28, respectively) was smaller on POD 14 than on POD 28 possibly due to a longer resting length (14.4 ± 3.6, 13.9 ± 1.8, 15.5 ± 1.7 mm on PODs 1, 14, and 28, respectively) on POD 28 than on POD 14. The nearly constant maximal shortening indicates that the muscle was unaffected by surgery. Resting length was systematically longer before maximal stimulation than during CO2 rebreathing. This may be due to the supine position for phrenic stimulation or to anesthesia abolishing active expulsion of increased ruminal volume due to fermentation of ruminal contents.

At autopsy, sonomicrometer crystal locations were not different radially between VATS and thoracotomy groups. The distal diaphragmatic crystal was located at 47.2 ± 5.3 and at 55.6 ± 10.4% of the average diaphragmatic length (103 ± 11 mm) from the costal edge for VATS and thoracotomy, respectively, whereas the proximal crystal was located at 27.3 ± 8.3 and 29.2 ± 6.5% of the average length from the central tendon for VATS and thoracotomy, respectively. Distance from the xiphoid to the line of sonomicrometer implantation on diaphragmatic muscle fibers was 27.1 ± 0.7% of 404.0 ± 30.1 mm and 41.8 ± 7.6% of 302.7 ± 16.2 mm of the right hemidiaphragmatic circumference for VATS and thoracotomy, respectively.

DISCUSSION

This study demonstrates that, although VATS and thoracotomy may differ, these differences are smaller than the group standard deviation. Thus we cannot support our hypothesis of a doubling of recovery speed for costal diaphragmatic function after VATS compared with thoracotomy. Although diaphragmatic function recovers within 3 wk of VATS, this 1-wk faster recovery is unlikely to explain the strong clinical impression that VATS is superior to thoracotomy in terms of recovery. Our sample size allowed us to detect shortening fraction differences between the early postoperative period and recovery in three of our four comparisons (VATS during QB and 10% CO2 rebreathing and thoracotomy during 10% CO2 rebreathing), confirming that four sheep would allow us to detect whether complete recovery occurred by 2 wk. Similarly, VT, VE, Pdi, and Pes generally were larger for the VATS group, further supporting our assertion that group differences in costal shortening fraction are clinically unimportant.

The inequality in resting length between VATS and thoracotomy enhanced their shortening fraction differences. Sonomicrometer focusing lenses occupy space in muscle and alter pulse transmission speed. Thus an LFRC difference would reduce shortening fraction of the closer sonomicrometers. We tested this by attaching sonomicrometers to movable wooden blades. They were immersed in water, separation distance was varied and measured, and the electronic output was recorded. We found a reported output of 2.2 mm when the lenses touched, and actual distance increased less than measured distance: actual distance (mm) = measured distance (mm) × 0.974 + 2.2 (R = 0.998).

When shortening fraction was recomputed from this relation, it increased for both groups, but more for thoracotomy, further reducing group differences. Thus shortening fraction differences favoring VATS decreased from an average of 2.4% LFRC before correction to 1.7% LFRC after correction during QB and from 1.4 to 1.5% during 10% CO2 rebreathing. Standard deviations increased. Thus smaller LFRC values for the thoracotomy group may have enhanced intergroup shortening fraction differences.

Causes for altered diaphragmatic contraction after thoracic surgery are incompletely understood but are likely to be similar after abdominal surgery. Reduced diaphragmatic shortening after abdominal surgery is attributed to depression of diaphragmatic activation (3), yet the responsible afferent pathways after thoracic or abdominal surgery remain to be elucidated, because anesthetic regimens incompletely reverse these changes (10, 19). Abdominal surgery produces a depression of diaphragmatic contraction (4), changes expiratory activation of external oblique and transversus abdominis muscles (6), and increases rib cage motion and reduces abdominal motion (8). Bilateral phrenic nerve blockade by local anesthesia produces similar effects (14, 26).

In contrast to the abdomen, where laparotomy exposes viscera equally, the mediastinum separates the thorax into two sides. Entering the right thorax via the right chest could leave the left thorax and hemidiaphragm unaffected. Similarly, unilateral phrenic nerve blockade produces less ventilatory impairment than bilateral blockade, being asymptomatic in humans (5, 17) or producing no respiratory distress in animals (14, 26). Our 54% depression of maximal diaphragmatic shortening (an overestimate, because a paralyzed diaphragm lengthens during inspiration) was mild and was associated with no parasternal muscle recruitment and small reductions of VT. However, if thoracic surgery only impairs diaphragmatic contraction on the side of surgery and by only slightly more than 50%, ventilatory changes could be relatively small, even without parasternal muscle recruitment. Indeed, if we assume that the right lung contributes 60% of a VT change and each hemidiaphragm produces 75% of its adjacent lung's VT (20), VT would be reduced an average of 11.9 ± 19.4% (SD) more than our observed VT values for PODs 1, 7, and 14 at both CO2 levels. This is within our experimental error and not different from zero. This discrepancy would be reduced by tidal rib cage expansion increasing in response to diaphragmatic inhibition, by diaphragmatic paralysis producing diaphragmatic lengthening rather than an isometric diaphragm, or by the contralateral diaphragm increasing its shortening.

If dysfunction affected only the right hemidiaphragm, Pga could be maintained during constant left diaphragmatic activation and contraction. Lisboa et al. (17) observed normal Pga changes (measured on the left side in the stomach) in patients with right hemidiaphragmatic paralysis. Similarly, Hillman and Finucane (12) reported a ratio of the change in Pga to the change in Pes in right-sided paralysis of -0.44 ± 0.42, which was significantly different from 0.67 ± 0.23 during left-sided paralysis. Thus, in our standing sheep, a right-sided diaphragmatic depression might not alter substantially tidal changes of inspired volume, Pes, or Pga.

Our lack of increased postoperative parasternal shortening, even if diaphragmatic contraction is altered unilaterally, is surprising. Katagiri et al. (14) noted a small but insignificant increase in parasternal shortening in dogs with unilateral diaphragmatic paralysis induced by local anesthetics. However, long-term parasternal sonomicrometer implantation may produce greater injury than long-term diaphragmatic implantation. Katagiri et al. reported successful measurements from only 40% of their parasternal sonomicrometers. Our POD -1 parasternal shortening measurements were markedly different between VATS and thoracotomy, and the VATS group declined rapidly while the thoracotomy group declined gradually. The dysfunction in each animal group may reflect the problems encountered by others (14). Indeed, examining some implantation sites at autopsy demonstrated changes consistent with muscle fibrosis and may explain our lack of parasternal muscle facilitation after surgery.

Relationships between intracavitary pressures, VT, and diaphragmatic contraction may be altered when diaphragmatic contraction is impaired unilaterally. Pga may reflect pressure under the left hemidiaphragm (17), whereas Pes will average hemithoracic changes (12, 13) or slightly emphasize right-sided changes (13). During right hemidiaphragm inhibition, pleural pressure generated by rib cage muscles and the normal left hemidiaphragm would become an afterload to the right hemidiaphragm. Whereas the right hemidiaphragm would contract, its generated tension and its shortening would be reduced disproportionately to VT (Fig. 2) and to Pdi (Fig. 3), since Pdi would appear minimally reduced and left-sided Pga would be nearly constant. Indeed, we previously described such uncoupling of global and regional measurements in patients undergoing a thoracic pulmonary resection. A postoperative thoracic epidural local anesthetic permitted stronger tidal Pes changes, did not alter diaphragmatic EMG activity, and converted small active diaphragmatic shortening to small paradoxical lengthening (10). This increased Pes, when diaphragmatic activation was constant, acted as a diaphragmatic afterload. Thus Pdi and Pes increased (tidal Pga was small and nearly constant), but active hemidiaphragmatic tension (as estimated by diaphragmatic EMG activity) remained constant, so diaphragmatic inspiratory length changes shifted from slight shortening to slight lengthening as afterload increased.

Our previous shortening measurements (9, 21, 22, 27, 28) are surprisingly similar to those in this study, when we consider that previous incisions extended from the sternum to the deep back muscles and that we incised the inferior pulmonary ligament. Thus about twice the skin, muscle, and pleura were incised and twice the intrathoracic surface was manipulated, but with minimal differences of diaphragmatic contractile depression. These observations suggest that inhibition is related to opening the chest and its intrathoracic manipulation rather than to the length of the chest wall or pleural incision or the diaphragmatic manipulation. In addition, because our previous studies compared the effects of interventions [digoxin (9), aminophylline (21), epidural anesthesia (22), or mechanical ventilation (27)] in the same group of animals, comparisons within the same subject would appear to be meaningful. However, we believe that detection of real differences between groups of different subjects receiving different treatments, e.g., surgical approach, might be obscured by differences caused by factors external to the true variable (29). However, diaphragm function was not altered substantially by the surgical approach in this sheep model of thoracic surgery.

ACKNOWLEDGEMENTS

The authors thank Warren M. Zapol for encouragement and support throughout the study, Alan Zaslavsky for many suggestions about statistical analysis, and Melahat Kavosi for excellent technical assistance.

FOOTNOTES

   This work is partly funded by Nellcor, Puritan Bennett Corp.

Address for reprint requests: W. R. Kimball, Dept. of Anesthesia, Massachusetts General Hospital, Boston, MA 02114.

Received 10 March 1997; accepted in final form 7 July 1997.

REFERENCES

1. Baydur, A., P. K. Behrakis, W. A. Zin, M. Jaeger, and J. Milic-Emili. A simple method for assessing the validity of the esophageal balloon technique. Am. Rev. Respir. Dis. 126: 788-791, 1982[Medline].
2. Bensard, D. D., R. C. McIntyre, B. J. Waring, and J. S. Simon. Comparison of video thoracoscopic lung biopsy to open lung biopsy in the diagnosis of interstitial lung disease. Chest 103: 765-770, 1993[Abstract/Free Full Text].
3. Dureuil, B., N. Viires, J. P. Cantineau, M. Aubier, and J. M. Desmonts. Diaphragmatic contractility after upper abdominal surgery. J. Appl. Physiol. 61: 1775-1780, 1986[Abstract/Free Full Text].
4. Easton, P. A., J. W. Fitting, R. Arnoux, A. Guerraty, and A. E. Grassino. Recovery of diaphragm function after laparotomy and chronic sonomicrometer implantation. J. Appl. Physiol. 66: 613-621, 1989[Abstract/Free Full Text].
5. Easton, P. A., J. A. Fleetham, A. De La Rocha, and N. R. Anthonisen. Respiratory function after paralysis of the right hemidiaphragm. Am. Rev. Respir. Dis. 127: 125-128, 1983[Medline].
6. Farkas, G. A., and A. De Troyer. Effects of midline laparotomy on expiratory muscle activation in anesthetized dogs. J. Appl. Physiol. 67: 599-605, 1989[Abstract/Free Full Text].
7. Ferson, P. F., R. J. Landreneau, R. D. Dowling, S. R. Hazelrigg, P. Ritter, S. Nunchuck, M. K. Perrino, C. M. Bowers, M. J. Mack, and M. J. Magee. Comparison of open versus thoracoscopic lung biopsy for diffuse infiltrative pulmonary disease. J. Thorac. Cardiovasc. Surg. 106: 194-199, 1993[Abstract].
8. Ford, G. T., W. A. Whitelaw, T. W. Rosenal, P. J. Cruse, and C. A. Guenter. Diaphragm function after upper abdominal surgery in humans. Am. Rev. Respir. Dis. 127: 431-436, 1983[Medline].
9. Fratacci, M. D., W. R. Kimball, J. C. Wain, E. M. Greene, and W. M. Zapol. Effects of digoxin on regional diaphragm function after thoracotomy in awake sheep. Respir. Physiol. 95: 259-271, 1994[Medline].
10. Fratacci, M. D., W. R. Kimball, J. C. Wain, R. M. Kacmarek, D. M. Polaner, and W. M. Zapol. Diaphragmatic shortening after thoracic surgery in humans: effects of mechanical ventilation and thoracic epidural anesthesia. Anesthesiology 79: 654-665, 1993[Medline].
11. Giudicelli, R., P. Thomas, T. Lonjon, J. Ragni, N. Morati, R. Ottomani, P. A. Fuentes, H. Shennib, and M. Noirclerc. Video-assisted minithoracotomy versus muscle-sparing thoracotomy for performing lobectomy. Ann. Thorac. Surg. 58: 712-718, 1994[Abstract].
12. Hillman, D. R., and K. E. Finucane. Respiratory pressure partitioning during quiet inspiration in unilateral and bilateral diaphragmatic weakness. Am. Rev. Respir. Dis. 137: 1401-1405, 1988[Medline].
13. Hurewitz, A. N., U. Sidhu, E. H. Bergofsky, and A. D. Chanana. How alterations in pleural pressure influence esophageal pressure. J. Appl. Physiol. 56: 1162-1169, 1984[Abstract/Free Full Text].
14. Katagiri, M., R. N. Young, R. S. Platt, T. M. Kieser, and P. A. Easton. Respiratory muscle compensation for unilateral or bilateral hemidiaphragm paralysis in awake canines. J. Appl. Physiol. 77: 1972-1982, 1994[Abstract/Free Full Text].
15. Kirby, T. J., M. J. Mack, R. J. Landreneau, and T. W. Rice. Lobectomy video-assisted thoracic surgery versus muscle-sparing thoracotomy: a randomized trial. J. Thorac. Cardiovasc. Surg. 109: 997-1002, 1995[Abstract].
16. Landreneau, R. J., S. R. Hazelrigg, M. J. Mack, R. D. Dowling, D. Burke, J. Gavlick, M. K. Perrino, P. S. Ritter, C. M. Bowers, J. DeFino, S. K. Nunchuck, J. Freeman, R. J. Keenan, and P. F. Ferson. Postoperative pain-related morbidity: video-assisted thoracic surgery versus thoracotomy. Ann. Thorac. Surg. 56: 1285-1289, 1993[Abstract].
17. Lisboa, C., P. D. Paré, J. Pertuze, G. Contreras, R. Moreno, S. Guillemi, and E. Cruz. Inspiratory muscle function in unilateral diaphragm paralysis. Am. Rev. Respir. Dis. 134: 488-492, 1986[Medline].
18. Maeda, H., K. Nakahara, K. Ohno, T. Kido, M. Ikeda, and Y. Kawashima. Diaphragm function after pulmonary resection: relationship to postoperative respiratory failure. Am. Rev. Respir. Dis. 137: 678-681, 1988[Medline].
19. Mankikian, B., J. P. Cantineau, M. Bertrand, E. Kieffer, R. Sartene, and P. Viars. Improvement of diaphragmatic function by a thoracic extradural block after upper abdominal surgery. Anesthesiology 68: 379-386, 1988. [Medline]
20. Mead, J., and S. H. Loring. Analysis of volume displacement and length changes of the diaphragm during breathing. J. Appl. Physiol. 53: 750-755, 1982[Abstract/Free Full Text].
21. Polaner, D. M., W. R. Kimball, M. D. Fratacci, J. C. Wain, A. Torres, R. M. Kacmarek, and W. M. Zapol. Effects of aminophylline on regional diaphragmatic shortening after thoracotomy in the awake lamb. Anesthesiology 77: 93-100, 1992[Medline].
22. Polaner, D. M., W. R. Kimball, M. D. Fratacci, J. C. Wain, and W. M. Zapol. Thoracic epidural anesthesia increases diaphragmatic shortening after thoracotomy in the awake lamb. Anesthesiology 79: 808-816, 1993[Medline].
23. Road, J., R. Vahi, P. DelRio, and A. Grassino. In vivo contractile properties of fatigued diaphragm. J. Appl. Physiol. 63: 471-478, 1987[Abstract/Free Full Text].
24. Road, J. D., K. R. Burgess, W. A. Whitelaw, and G. T. Ford. Diaphragm function and respiratory response after upper abdominal surgery in dogs. J. Appl. Physiol. 57: 576-582, 1984[Abstract/Free Full Text].
25. Simonneau, G., A. Vivien, R. Sartene, F. Kunstlinger, K. Samii, Y. Noviant, and P. Duroux. Diaphragm dysfunction induced by upper abdominal surgery: role of postoperative pain. Am. Rev. Respir. Dis. 128: 899-903, 1983[Medline].
26. Stradling, J. R., L. F. Kozar, J. Dark, T. Kirby, S. M. Andrey, and E. A. Phillipson. Effect of acute diaphragm paralysis on ventilation in awake and sleeping dogs. Am. Rev. Respir. Dis. 136: 633-637, 1987[Medline].
27. Torres, A., R. M. Kacmarek, W. R. Kimball, J. Qvist, K. Stanek, R. Whyte, and W. M. Zapol. Regional diaphragmatic length and EMG activity during inspiratory pressure support and CPAP in awake sheep. J. Appl. Physiol. 74: 695-703, 1993[Abstract/Free Full Text].
28. Torres, A., W. R. Kimball, J. Qvist, K. Stanek, R. M. Kacmarek, R. I. Whyte, G. Montalescot, and W. M. Zapol. Sonomicrometric regional diaphragmatic shortening in awake sheep after thoracic surgery. J. Appl. Physiol. 67: 2357-2368, 1989[Abstract/Free Full Text].
29. Wakai, Y., A. M. Leevers, and J. D. Road. Regional diaphragm shortening measured by sonimicrometry. J. Appl. Physiol. 77: 2791-2796, 1994[Abstract/Free Full Text].
30. Waller, D. A., J. Forty, and G. N. Morritt. Video-assisted thoracoscopic surgery versus thoracotomy for spontaneous pneumothorax. Ann. Thorac. Surg. 58: 372-377, 1994[Abstract].
0161-7567/97 $5.00 Copyright © 1997 the American Physiological Society


This article has been cited by other articles:


Home page
 Anesth. Analg.Home page
C. Weissman
Pulmonary Function After Cardiac and Thoracic Surgery
Anesth. Analg., June 1, 1999; 88(6): 1272 - 1272.
[Full Text] [PDF]

Home page
 ThoraxHome page
N M Siafakas, I Mitrouska, D Bouros, and D Georgopoulos
Surgery and the respiratory muscles
Thorax, May 1, 1999; 54(5): 458 - 465.
[Full Text]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Imanaka, H.
Right arrow Articles by Kacmarek, R. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Imanaka, H.
Right arrow Articles by Kacmarek, R. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online