Methods

This randomized 2-center study included patients admitted to the coronary care unit for STEMI without signs of acute heart failure on admission, i.e., with no dyspnea, at Killip class I with a normal chest x-ray, and without previous heart failure. During hospitalization, repeated lung auscultations were done, and respiratory rate, heart rate, blood pressure, and oxygen saturation were recorded every hour. Diagnosis of PED was based on chest x-rays and a modified Killip classification: class I, no dyspnea and no rales; class IIA, no dyspnea but detection of rales at the lung bases; class IIB, no dyspnea and rales over the lower 1/2 of the lung fields; class IIC, dyspnea and rales over the lower 1/2 of the lung fields; class III, dyspnea and rales over most or the entire lung fields; and class IV, cardiogenic shock. The study protocol was approved by the local institutional review board and the ministry of health ethics committee (HT3-657-1). Participants provided signed informed consent.

LI was monitored in this study using a noninvasive impedance device (RSMM Company, Tel Aviv, Israel) as described previously and measured in ohms but reported as percent change from initial measurement. Such an LI presentation with patients serving as their own controls is widely used and allows the assessment of changes in lung fluid content but not of the absolute value of lung fluid volume. An LI change of 1 Ω is usually equivalent to a 1.5% to 2.5% change from baseline value. Regular electrocardiographic electrodes were used for LI monitoring with measurements at 60-minute intervals. Chest x-ray was used to validate changes in lung fluid content. We formulated a radiologic score (RS) for quantitative assessment of roentgenologic signs of lung congestion. The RS was calculated as the sum of sign values and defined as follows: 0 to 1, no lung edema; 2 to 4, interstitial lung congestion; 5 to 6, mild alveolar edema; 7 to 8, moderate alveolar edema; and 9 to 10, severe alveolar edema.

Group 1 included all study patients whose maximal LI decrease during monitoring did not exceed 12%. The remaining patients were randomized on reaching the LI threshold decrease of 12% to 14% to a conventional treatment (group 2) or to pre-emptive LI-guided therapy (intervention group, group 3) at a 2:1 ratio. Study patients were treated for STEMI according to current guidelines. In group 2, initiation of treatment for PED was based on clinical assessment (patient report of dyspnea and 1 of the following: lung rales or signs of congestion or edema on chest x-ray). Medical therapy included an intravenous furosemide bolus of 40 mg and a continuous furosemide drip of 125 mg/day followed by an oral furosemide dose of 40 to 80 mg/day. Duration of furosemide treatment was determined by the attending physician based on patient assessment. After randomization, physicians treating group 2 patients were blinded for LI values. In group 3, pre-emptive treatment was begun immediately after randomization. The treatment protocol used in group 3 was identical to that in group 2 except that furosemide treatment was continued until the return of LI to the initial value. Physicians treating these patients were unblinded to LI values. If treatment was not satisfactory, then the diuretic dosage could be changed at the discretion of attending physician. Recommendations for treatment at discharge were provided according to guidelines. After discharge, patients were followed and treated by family physicians. For assessment of long-term outcome, hospital records and computerized clinic databases were reviewed and 6-month telephone contacts with patients were done.

Continuous data were presented as mean ± SD or median (interquartile range). The LI values corresponding to the transition zone from the interstitial to the alveolar stages of pulmonary congestion were assessed by receiver operating curve analysis. Categorical data were presented as count or proportion (percentage). Comparisons among multiple parameters within the same group or different groups were carried out using 1-way analysis of variance. Pearson coefficient was calculated for correlation analysis. Fisher’s exact test and chi-square test were used to detect differences in the prevalence of categorical variables (death and PED) among the 3 groups. Analysis of variance and Kruskal–Wallis with post hoc test were applied to compare differences in hospital length of stay and peak creatine kinase levels among groups. Multivariate analysis by logistic regression was employed to study parameters associated with length of stay, rehospitalizations, death, and occurrence of new-onset heart failure. Kaplan–Meier survival curves were drawn to show time to death. A p value <0.05 was considered statistically significant. SPSS 18.0 (SPSS, Inc., Chicago, Illinois) was used.

Results

In the preliminary phase of the study, we evaluated 65 patients with STEMI (60 ± 12 years old) who had no clinical (Killip class I) or radiographic signs of PED at admission. During 72 hours of monitoring 33 patients with STEMI did not develop clinical and radiologic signs of PED and another 32 well-matched monitored patients developed PED (Killip classes IIA to IV). LI decreased in the former group during hospitalization by 0% to 14% from the initial value and in the latter group by 12% to 27% at the time rales were detected at the lung bases for the first time. Receiver operating curve analysis to discriminate between these groups yielded areas under the curve equal to 0.98 for an LI decrease of 12%, 0.997 for an LI decrease of 13%, 0.996 for 14%, and 0.971 for 15%. Therefore, we selected the LI decrease range 12% to 14% as representing the onset of the transition zone from interstitial to alveolar edema. Attaining this threshold portends evolution to PED with a sensitivity of 96% to 100% and a specificity of 99% to 100%.

For the present study we screened 689 consecutive patients with STEMI without dyspnea. Of these, 77 patients demonstrated signs of heart failure (Killip classes IIA to IIB) on physical examination, and an additional 52 asymptomatic patients (Killip class I) showed radiologic signs of PED on admission. These 129 patients were excluded, and the remaining 560 patients were enrolled and monitored for 84 ± 36 hours. Clinical and laboratory parameters of study groups are presented in Table 1 .

Of the 560 study patients, 347 patients (62%) demonstrated an LI decrease ≤12%. These patients did not report dyspnea and remained at Killip class I during the entire monitoring period (group 1). LI changed during monitoring but maximal LI decrease was only 5.8% (95% confidence interval −0.4 to −12.0, p <0.001 vs initial value). The RS at this time (1.9 ± 1.2) was compatible with interstitial edema and correlated with LI (r = −0.68, p <0.001). Changes of respiratory rate, heart rate, or oxygen saturation from baseline during the follow-up period were not significant.

Time that elapsed from the beginning of monitoring to the point LI reached the threshold decrease of 12% to 14% in the 213 patients was 9 ± 7.2 hours (median 7.3, interquartile range 4 to 13). Corresponding RS increased from 0.3 ± 0.4 to 2.8 ± 1.2 (interstitial edema, p <0.01). At this stage, although patients did not yet complain of dyspnea (Killip class I), they were randomized to conventional treatment (n = 142, group 2) and to LI-guided pre-emptive treatment (n = 71, group 3). LI changes and their time of occurrence throughout the monitoring period in groups 2 and 3 are presented in Figure 1 . Time for any event or measurement (i.e., reaching LI threshold level, first detection of rales, etc.) was calculated by averaging that particular period for all patients.

Dynamics of lung impedance in groups 2 (dotted bars) and 3 (hatched bars) by events (A) Baseline lung impedance in groups 2 and 3. (B) Lung impedance decrease by 12% to 14%; randomization of patients to groups 2 and 3. Patients in group 2 progressed through the (C) appearance of lung rales, (D) onset of dyspnea, (E) moderate and (F) severe pulmonary edema, and (G) final recovery. Patients in group 3 showed (H) maximal decrease of lung impedance in patients in whom pre-emptive treatment prevented pulmonary edema, (I) maximal decrease of lung impedance in patients in whom pre-emptive treatment did not prevent pulmonary edema, and (J) lung impedance at end of monitoring. Follow-up in hours is shown (x axis) .

Four hours after randomization (interquartile range 1 to 7), basilar lung rales were first detected (Killip class IIA) in the still asymptomatic group 2, with LI decreasing by 18% (95% confidence interval 14 to 23) and RS increasing to 4.5 (p <0.001 vs baseline). No significant change in clinical parameters was observed at this point (increase in respiratory rate and heart rate and decrease in oxygen saturation, p = NS). Shortly after, at a median of 5.4 hours (interquartile range 1 to 8), patients began to report shortness of breath, with LI decreasing by 26 ± 4% and RS increasing to 5.6 ± 0.9 (p <0.001). Respiratory rate and heart rate increased (p <0.01) and oxygen saturation decreased (p <0.01) and treatment was initiated. Despite treatment, PED progressed with further LI decrease and increasing corresponding RSs (p <0.001; Figure 1 ). In 132 patients (93%) progression of PED was halted at Killip classes IIC to III, whereas in 10 patients (7%) it progressed to cardiogenic shock (Killip class IV), with 6 patients (4%) dying owing to pump failure. At hospital discharge, LI and RS significantly improved in surviving patients (p <0.01; Figure 1 ).

LI-guided pre-emptive treatment was begun in group 3 in the asymptomatic stage (Killip class I) immediately after randomization at an LI decrease of 13.8 ± 0.6% from baseline (p <0.001; Figure 1 ). LI decrease was halted after treatment at an LI decrease of 17 ± 2.6% (p <0.01) in 63 patients (89%). These patients remained asymptomatic. Only 8 patients in group 3 (11%) developed clinically overt PED (Killip classes IIC to III) with an LI decrease of 25 ± 4% (p <0.01; Figure 1 ). While on treatment, they achieved an LI return to baseline value. The proportion of patients in group 3 who developed PED (11%) was smaller than that in group 2 (100%, p <0.0001). No patient in group 3 progressed to cardiogenic shock. Compared to group 2, no death occurred in group 1 or 3 during hospitalization (p <0.01). LI correlated with RS (r = −0.9, p <0.01) in groups 2 and 3. A lower furosemide dosage was used in group 3 than in group 2 (p <0.001; Table 2 ). Discharge medications were based on current guidelines as presented in Table 2 .

Lengths of stay in the coronary care unit and in the hospital are presented in Table 2 . After adjustment for variables, we found that group 2 required a longer length of stay in the hospital compared to groups 1 and 3 (p <0.0001; Table 3 ).

Table 3

Multivariate logistic regression analysis adjusted for age, left ventricular ejection fraction, cardiovascular risk factors, maximal creatine kinase, and admission creatinine and hemoglobin on hospital length of stay for rehospitalization rate during first year after discharge, for six-year occurrence of new heart failure, and on six-year mortality

Variables	LOS >5 Days		1-Year Rehospitalization		6-Year New HF		6-Year Mortality
	OR (95% CI)	p Value	OR (95% CI)	p Value	OR (95% CI)	p Value	OR (95% CI)	p Value
Age ⁎	1.03 (1.01–1.05)	0.013	1.02 (0.99–1.04)	0.063	1.03 (1.001–1.06)	0.013	1.04 (1.01–1.07)	0.008
Left ventricular ejection fraction (%) †
≤30	ND	0.66	ND	0.81	19.70 (6.43–60.2)	<0.0001	7.48 (2.52–22.1)	<0.0001
31–40					3.36 (1.41–8.05)	0.006	1.93 (0.71–5.25)	0.19
42–50					1.17 (0.47–2.92)	0.74	2.21 (0.91–5.38)	0.08
Maximal creatine kinase ‡
Quartile 2 388–1,100	0.99 (0.52–1.89)	0.98	ND	0.64	ND	0.86	ND	0.11
Quartile 3	1.59 (0.80–3.16)	0.19
Quartile >2,216	2.82 (1.42–5.60)	0.003
Acute myocardial infarction localization (anterior vs other)	1.53 (0.94–2.49)	0.08	1.1 (0.69–1.72)	0.7	1.10 (0.54–2.21)	0.80	0.92 (0.45–1.90)	0.83
Diabetes mellitus	1.24 (0.71–2.16)	0.44	1.84 (1.09–3.09)	0.022	2.34 (1.13–4.85)	0.022	2.56 (1.16–5.67)	0.02
Hypertension	0.76 (0.47–1.23)	0.27	1.19 (0.77–1.86)	0.44	2.31 (0.66–2.59)	0.44	1.16 (0.56–2.38)	0.7
Hyperlipidemia	0.75 (0.45–1.27)	0.29	0.63 (0.38–1.03)	0.065	0.60 (0.29–1.24)	0.17	0.51 (0.23–1.11)	0.09
Current smoking	0.71 (0.44–1.15)	0.17	1.2 (0.76–1.88)	0.44	0.97 (0.48–1.95)	0.93	1.11 (0.53–2.32)	0.78
Admission creatine (change +0.1 mg/dl)	1.06 (0.54–2.09)	0.06	0.93 (0.49–1.81)	0.84	0.66 (0.26–1.74)	0.42	1.39 (0.59–3.25)	0.45
Admission hemoglobin (change +0.1 g/dl)	0.90 (0.78–1.05)	0.19	0.98 (0.85–1.12)	0.72	1.13 (0.92–1.38)	0.24	0.96 (0.78–1.19)	0.73
Adjustment for these variables compared to group 1
Group 2	5.28 (3.10–8.99)	<0.0001	3.62 (2.16–6.06)	<0.0001	15.7 (6.15–40.1)	<0.0001	3.46 (1.56–7.69)	0.002
Group 3	1.40 (0.73–2.68)	0.31	1.78 (0.96–3.29)	0.066	4.52 (1.55–13.17)	0.006	1.0 (0.32–3.16)	1.0
Adjustment for these parameters compared to group 3
Group 2					3.41 (1.56–7.48)	0.002	3.2 (1.13–9.01)	0.028

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. A Computed Tomography-Based Coronary Lesion Score to Predict Acute Coronary Syndrome Among Patients With Acute Chest Pain and Significant Coronary Stenosis on Coronary Computed Tomographic Angiogram
2. Role of CHADS 2 Score in Evaluation of Thromboembolic Risk and Mortality in Patients With Atrial Fibrillation Undergoing Direct Current Cardioversion (from the ACUTE Trial Substudy)
3. Effect of Bisoprolol on Respiratory Function and Exercise Capacity in Chronic Obstructive Pulmonary Disease
4. Usefulness of Circulating Biomarkers for the Prediction of Left Ventricular Remodeling After Myocardial Infarction

Stay updated, free articles. Join our Telegram channel

Join