We studied 32 patients (mean age 62 years; 8 female) having elective lung resection between January 1 ^st and September 30 ^th 2000. IRB approval and written informed consent were obtained. All patients underwent either fiberoptic or rigid bronchoscopy and had histology samples obtained preoperatively. Routine check-up consisted of thoracic imaging by standard chest x-rays and CT scans. Staging was completed by abdominal ultrasonography and cerebral CT scans. Where necessary, bone scans as well as ventilation and perfusion scintigraphies were performed for exact preoperative evaluation.

Inclusion criteria were patients suitable for surgery with pre- or intraoperatively staged non small cell lung carcinoma (NSCLC, n=28) with N0 or N1 disease on CT scan, biopsy bronchoscopy or mediastinoscopy. After surgery, by final pathology report, 4 cases turned out to have different types of histology (1 carcinoid, 1 SCLC “limited disease”, 1 benign lesion, 1 metastasis). 16 patients underwent lobectomies (11 upper lobe, 11 right sided), 3 sleeve lobectomies, 10 pneumonectomies (5 right sided) and 3 sleeve pneumonectomies. No preoperative chemo- or radiotherapy was performed in these patients.

Exclusion criteria were higher grades of pulmonary, cardiac, renal or arteriosclerotic diseases, aortic aneurysm and general contraindications against general anaesthesia or thoracotomy, as well as other malignant diseases. Vital capacity or FEV1 of less than 60% predicted value lead to exclusion.

After induction of general anaesthesia by standardised means with electrocardiography, oxygen saturation and blood pressure monitoring using total intravenous anaesthesia (TIVA), a standard central venous line (Arrow™ International, Reading, Pennsylvania, USA) was introduced via the ipsilateral jugular vein. Central venous pressure was recorded. All patients were extubated immediately after surgery. Patients were routinely transferred from the ICU to regular wards after 48 hours.

There has so far been a general recommendation against using the single indicator thermodilution method (PiCCO™) in thoracic surgery. It was thought that pulmonary blood volume may be altered by the operation and thus lead to inexact values [₁₀], because PiCCO™ does not measure pulmonary blood volume index (PBVI), but adds constantly 25% to global enddiastolic volume index (GEDI) ₁ . In double indicator method (COLD™), the data achieved are actual measured parameters. The indocyanin green is protein bound and stays strictly in the intravascular space, thus reflects unaltered data even during possible changes in factors like pulmonary blood volume. Since the double indicator method is a more invasive (large arterial introducer needed), risky (protein bound indicator may cause allergies and shock), technically more difficult and time consuming and, finally, more costly, the goal was to establish a single indicator method useful in lung resections. Agreement between PiCCO™ and COLD™ concerning cardiac index (CI) and GEDI was well investigated in the development of the PiCCO™-system [₁,₁₁,₁₂, ₂₄, ₂₅ ]. For comparison of ITBI and ELWI the first 9 patients underwent both double and single indicator methods.

Double indicator method (n=9) (COLD™): A fiberoptic thermistor tipped catheter (Pulsiocath™, PV2023, Pulsion Medical Systems, Munich, Germany) was placed via the ipsilateral femoral artery by a 4 French introducer and positioned in the infrarenal aorta. The catheter was then connected to a double indicator computer device (COLD™ Z-021, Pulsion Medical Systems, Munich, Germany). In serial a single indicator computer device (PiCCO™, Pulsion Medical Systems, Munich, Germany) was connected. 10 ml of cold indocyanin green (ICG-Pulsion™, Pulsion Medical Systems, Munich, Germany) mixed with cold (less than 8° Celsius) normal saline were administered into the central venous line for the parallel measured of double and single indicator method.

Single indicator method (n=23) (PiCCO™): A thermodilution catheter (Pulsiocath™, PV2014L08, Pulsion Medical Systems, Munich, Germany) was placed directly in the ipsilateral femoral artery. 20 ml of cold (less than 8° Celsius) normal saline were administered into the central venous line for single indicator thermodilution method.

Cardiac index (CI), intrathoracic blood volume (ITBI) and extravascular lung water index (ELWI) were recorded [₁,₁₁,₁₂,₂₄, ₂₅ ]. To obtain more exact values than shown on PiCCO™, with no decimals, we calculated the absolute EVLW values by dividing it by the patient's weight in kilograms, obtaining a calculated ELWI. The PiCCO™ device calculated these data by mean transit time of the thermal indicator. ITBI is calculated by global enddiastolic volume index (GEDI) plus pulmonary blood volume index (PBVI). ELWI is intrathoracic thermo volume (“needle-to-needle”) minus ITBI.

To receive the relative changes of ITBI and ELWI compared to the preoperative situations, the preoperative value for ITBI and ELWI were set to 100%. The postoperative change was given in percent changes at 0, 12, 24, 36 and 48 hours postoperatively for each single patient.

To account for the resected lung volume in thoracic surgery procedures we corrected the ELWI-relative to the predicted remaining lung volume following surgery. Based on anatomical data [₁₃], depicted in Table 1, the ELWI-relative was transformed to ELWI-corrected using the term: ELWI-corrected (%) = ELWI-relative (%) / remaining lung volume (%) x 100.

Figure 1

Table 1: Estimated remaining lung volume in % following anatomical resection.

After placing the patient into the contralateral side position, a standard posterolateral thoracotomy was performed and radical excision of tumorous tisue by lobectomy or pneumonectomy. Patients were randomized in two groups at the time of data acquisition concerning the type of lymph node dissection: the “sampling” group with mediastinal lymph node sampling and the “radical” group with radical mediastinal lymphadenectomy by surgeons preference. Three patients received a bronchus sleeve procedure and 3 patients a sleeve pneumonectomy in combination with radical mediastinal lymphadenectomy in all 6 patients. At time of data analysis the results of these sleeve-resected patients were different from the non-sleeve resected patients. A third group “radical with sleeve resection” was created by these patients data.

Unless otherwise mentioned, all data are presented as means and standard error of the mean (SEM) for all variables. Continuous variables were checked for normality by plotting histograms. Variables that were not normally distributed were analysed using the Kruskal-Wallis test and the Mann-Whitney U test. Those that were normally distributed were assessed with a one-way analysis of variance and the Student's t-test. Chi-square test for trend was used for non-normally distributed variables and Repeated Measures was used for normally distributed values and their changes over time. To prove agreement of ITBI and ELWI between COLD™ (a) and PiCCO™ (b) system, the Bland-Altman test was used requiring >95% of all measurement t differences to be inside the two standard deviation (SD) interval [₁₄]. P values less than 0.05 were considered statistically significant. All tests were two-tailed. All statistical analyses were performed on a personal computer with the statistical package SPSS™ for Windows (Version 10.0, SPSS™, Chicago, USA).

Nine patients underwent simultaneous double and single indicator method (COLD™ (a) and PiCCO™ (b)) to determine the degree of agreement between these systems. Bland-Altman testing showed significant agreement (>95% value differences within two SD) for all observed parameters, confirming prior studies in patients not undergoing lung resection surgery [₁,₁₁,₁₂, ₂₄, ₂₅ ]. Lung surgery related results for ITBI and ELWI are shown in Figure 1, previously published, cardiac related parameters are not shown (CI, GEDI). We concluded that single indicator method (PiCCO™) delivers valuable data compared with the double indicator method (COLD™). Therefore the single indicator method alone was used in the remaining 23 patients.

Figure 2

Figure 1: Agreement between double and single indicator method (COLD™ and PiCCO™) Bland-Altman testing showed significant agreement (>95% value differences within two SD, n=48) for ITBI [A] and ELWI [B] between double (a=COLD™) and single (b=PiCCO™) indicator method systems (ITBI (COLD™-PiCCO™) mean 23.1 ± 96.4 SD ml/m ; ELWI (COLD™-PiCCO™) mean –0.21 ± 1.05 SD ml/kg).

The “sampling” group included 17 patients; the “radical” group -including the sleeve procedures- had 15 patients. No significant differences were seen between these groups concerning mean age 63 (52-75) years for the “sampling” group and 61 (41-72) years for the “radical” group. Vital capacity (92.3 ±13.3% SD versus 92.9 ±17.14% SD), FEV1 (84,6 ±12.9% SD versus 83.0 ±14.8% SD) pO₂, pCO₂, and SaO₂ were not significantly different between groups and performed procedures. Changes in CI, GEDI and CVP were not significantly different between study groups (sampling versus radical) and performed procedures throughout the observation period.

Looking at differences between performed procedures (lobectomies, n=19; pneumonectomies, n=13) independently of the type of lymph node dissection, there were significant (p=0.032) changes throughout the 48 hours period postoperatively concerning the relative ITBI including the sleeve procedures. The values for pneumonectomies were predominately lower throughout the observation period (Figure 2A) showing that these cases were consistently carried out with less volume substitution. The relative ELWI decrease was procedure specific, with a mean decrease of 11.7% seen after lobectomy and 33.6% after pneumonectomy (p=0.006). There were no statistically significant changes found throughout the postoperative observation period within the same procedure group (Figure 2B).

Figure 3

Figure 2: ITBI relative and ELWI relative differances by type of lung resection

[A] Relative ITBI shown for all cases regardless of type of lymph node dissection. Throughout the study period, pneumonectomies showed significantly lower values for ITBI relative compared to lobectomies (p=0.032), reflecting the reduced volume substitution in pneumonectomy procedures. Relative ELWI significantly decreased following lung surgery by the amount of resected lung tissue (lobectomies ~11.7%, pneumonectomies ~33.6%; p=0.006). Throughout the observation, no statistical significant changes were observed within the procedure groups.

Since the step down decrease of ELWI-relative following surgery reflected the degree of lung tissue resection, ELWI was corrected by the anatomical lung volume resection (Table 1).

No significant changes (p>0,3) in ITBI relative and ELWI corrected were observed throughout the 48 hours study period in lobectomies (Figure 3A+B) and pneumonectomies (Figure 4A+B).

Figure 4

Figure 3: Lobectomies: ITBI relative and ELWI corrected

[A] Relative ITBI was not different between sampling or radical lymphnode dissection and radical with sleeve resection lobectomies. Corrected ELWI was significantly increased in sleeve resected cases within the first 12 hours of observation compared to non sleeve procedures (p=0.003). No significant changes in corrected ELWI were observed between sampling and radical lymphnode dissection.

Figure 5

Figure 4: Pneumonectomies: ITBI relative and ELWI corrected

[A] Relative ITBI was significantly decreased throughout the observation period of 48 hours in sleeve pneumonectomies (p=0.002) compared to non-sleeve procedures. Corrected ELWI increased in all pneumonectomy cases after surgery. Sleeve resected cases peeked at a mean of 175 ±21% normalized to the preoperative value (100%), resulting in a significant increase with in the first 12 hours after sleeve pneumonectomy (p=0.031) compared to non-sleeve procedures. Between 24 and 48 hours following pneumonectomy a slight increase in ELWI corr. was observed in all groups, showing no significant differences between groups. Comparing early ITBI and ELWI results, sleeve pneumonectomies showed a high potential to accumulate lung water even in a setting of significantly lower ITBI.

In sleeve lobectomies the relative ITBI was not different compared to non sleeve lobectomies (Figure 3A) and ranged in the mean between 93-116% of the preoperative value. A significant increase in corrected ELWI was observed in sleeve resected patients with in the first 12 hours of observation compared to non sleeve procedures (sampling and radical combined p=0.003). The values for lung volume corrected ELWI in the sleeve resected group was found to be a mean of 169 ±47% postoperatively and 175 ±40% at 12 hours, a significant increase compared to the values in the non sleeve resected group, which ranged between 105-133% (Figure 3B).

The findings in sleeve pneumonectomies were similar to the findings in sleeve lobecomies. Throughout the observation period of 48 hours, the relative ITBI was significantly decreased in patients who underwent sleeve pneumonectomies (p=0.002). ITBI relative values ranged between 59-85% of the preoperatively measurements, where as non sleeve resected patients showed stable postoperative values within the first 24 hours of observation (95-106%) and even slight increases in relative ITBI for the following time period (100-116%) (Figure 4A). The corrected ELWI increased in all pneumonectomy cases after surgery. For non sleeve resected cases, an increase of the mean value to 130-136% was observed. Compared to these values, sleeve resected cases peeked at a mean of 175 ±21% normalized to the preoperative value (100%). This extreme rise of corrected ELWI within the first 12 hours after pneumonectomy in sleeve resected cases was significant compared to the non sleeve resected cases (p=0.031). Between 24 and 48 hours following pneumonectomy a slight increase was observed in all groups, with mean ranges of 125-148%, showing no significant differences (Figure 4B). A summary observation comparing early ITBI and ELWI results in Figure 4 showed the high potential in sleeve pneumonectomies to accumulate lung water in a setting of even significant lower intrathoracic blood volume.