# The Effect of Intermittent. Negative Pressure Ventilation and Long-term Oxygen Therapy for Patients With COPD – Results

Table 1 shows the anthropometric data, the blood gas analysis measurements, and the spirometric values obtained from the two groups of patients. These data indicate that group A presented a more marked degree of bronchial obstruction and a more severe hypoxemia and hypercapnia at discharge.

Thirty patients (86 percent) of group A and 32 patients (73 percent) of group В presented relapses of ARF during the follow-up with a total number of 1 to 11 episodes per patient in group A (mean ± SD = 3.31 ± 2.86) and 1 to 8 episodes per patient in group В (mean±SD = 1.95 ±2.1). Each episode was treated with our conservative ventilatory method. The values of RM of the two groups were similar (RM = 0.09 ± 0.08 in group A, RM = 0.13 ± 0.22 in group B). The disease-free interval between the beginning of follow-up and the first relapse was 10.616.5 and 12.9 ±12.1 months in groups A and B, respectively. Twenty-one (48 percent) patients of group В and 20 (57 percent) of group A had a relapse of ARF during the first 12 months of follow-up.

The mean rate of decline of Pa02 over time was 1.58 ± 5.59 mm Hg/yr in group A and 4.70 ± 8.25 mm Hg/yr in group В with a statistical difference at limits of significance (t= 1.92, p = 0.059). The rate of increase of PaC02 over time was significandy lower in group A (1.08 ± 3.41 mm Hg/yr) than in group В (4.91 ± 7.14 mm Hg/yr; t = 2.92; p = 0.005). The individual values of these two rates (slopes of the regression lines) were included in the Cox statistics described below. buy birth control

The actuarial survival curves of the two groups are shown in Figure 1. The total number of deaths was 17 in group A and 28 in group B. The F test showed that the difference in survival between groups A and В was significant (p = 0.05, F = 1.65 with 32,56 df).

The multivariate analysis of factors influencing survival is reported in Table 2. This analysis included both variables at entry and variables obtained during the follow-up. The results show that the values of time-trend of PaC02, Pa02 at discharge, RM, and FEV/ VC were significant prognostic indices. (In a preliminary univariate analysis, also the presence of cor pulmonale was shown to be a significant factor; however, in the multivariate analysis, this variable failed to reach significance probably because it was highly correlated with FEVj/VC.) The administration of LTOT was a factor influencing survival at levels of statistical significance.

By definition, most of the above-mentioned variables are devoid of predictive value because they were calculated retrospectively from follow-up information. Hence, the multivariate statistics was rerun using the subset of variables available at the beginning of the follow-up (namely, Pa02, PaC02, FEVv FEV/VC, presence of cor pulmonale, and hematocrit). Among these, only the FEV/VC was found to be below the statistical level of p = 0.10 (relative risk for each 1 percent increase in the index = 0.97 with 95 percent confidence interval of 0.94 to 0.99, p = 0.022).

**Table 1—Anthropometric Data, Blood Gas Analysis Measurements, Spirometric Data, and Cor Pulmonale Incidence in the 79 Patients at Beginning of the Follow-up**

Group A (n = 35) | Group В (n = 44) | StatisticalSignificance! | |

Age, yr | 65.8±7.8 | 63.6±8.7 | NS |

Sex, M/F | 26/9 | 37П | NS |

PaO_{a}, mm Hg |
|||

On admission | 38.0± 11.8 | 36.0 ±10.0 | NS |

At discharge | 51.8±5.7 | 61.6±11.3 | p<0.01 |

PaCO*, mm Hg | |||

On admission | 77.0± 12.8 | 83.0±15.5 | NS |

At discharge | 54.4 ±6.0 | 48.2 ±6.6 | p<0.01 |

pH | |||

On admission | 7.30 ±0.06 | 7.26±0.09 | NS |

At discharge | 7.37 ±0.03 | 7.38±0.04 | NS |

FEV^ %t$ | 24.3 ±9.9 | 31.0±11.7 | p<0.05 |

Щ %t$ | 211 ±43 | 205 ±40 | NS |

FEV/FX % | 34 ±10 | 40 ± 11 | p<0.05 |

Cor pulmonale | 19 (54%) | 27 (61%) | NS |

**Table 2—Multivariate Cox Analysis cf Variables Influencing Survival at p Level cf 0.10 or Less in 79 Patients With COPD Followed Up Over 4 Years After an Episode of Acute Respiratory Failure**

Variables* | StatisticalSignificancet | RelativeRisk |

Tune-trend of PaCO*: increase of <3 mm Hg/yr or no increase increase of ^3 mm Hg/yr | p = 0.0001 | 1 |

4.21 (2.02-8.78) | ||

Pa0_{2} at discharge, mm Hg |
p = 0.0002 | |

<60 | 1 | |

>60 | 0.23 (0.11-0.50) | |

Administration of LTOT: | p = 0.0006 | |

Yes | 1 | |

No | 3.32 (1.67-6.60) | |

RM | p = 0.0075 | |

<0.1 | 1 | |

s>0.1 | 2.53 (1.28-5.00) | |

FEV/VC: | p = 0.015 | |

For each 1% increase | 0.96 (0.93-0.99) |

*Figure 1. Survival of the two patient groups (solid line = group A; dotted line = group B).*