Measurement of the approximate elapsed time of ventricular fibrillation and monitoring the response of the heart to therapy

• US 5,077,667 A
• Filed: 06/13/1990
• Issued: 12/31/1991
• Est. Priority Date: 07/10/1989
• Status: Expired due to Term

• Current Assignee: Ohio State University The A Branch Government of Oh
• Original Assignee: Ohio State University
• Inventors: Dzwonczyk, Roger, Brown, Charles G.
• Primary Examiner(s): Smith, Jerry
• Assistant Examiner(s): Huntley, David
• Application Number:
  US07/537,365
• Time in Patent Office: 566 Days
• Field of Search: 364/413.05, 364/413.06, 128/702, 128/705, 128/419 D
• US Class Current: 607/5
• CPC Class Codes: A61B 5/361 Detecting fibrillation
• CPC Class Codes: A61B 5/4839 combined with drug delivery
• CPC Class Codes: A61B 5/7253 characterised by using transforms
• CPC Class Codes: A61B 5/7257 using Fourier transforms

Abstract

The approximate elapsed time since the onset of ventricular fibrillation is detected from an analog electrocardiogram signal. The signal is digitized for a time interval of four seconds to obtain a data set of time domain samples. These time domain samples are Fourier transformed to a frequency domain spectrum and the median frequency which bisects the energy of the power spectrum is detected. That median frequency is then compared to a pattern of experimentally obtained median frequency data as represented by a mathematical algorithm to calculate the estimated time from the onset of ventricular fibrillation. This frequency parameter can also be used to evaluate the response to therapy during ventricular fibrillation and CPR, as well as estimate the most appropriate time to defibrillate a subject following various pharmacologic and mechanical intervention.

Claims

1. A method for measuring a clinically useful characteristic of the fibrillating heart of a subject which correlates with the approximate elapsed time since the onset of ventricular fibrillation of the heart, and with the metabolic state of the heart after therapeutic intervention, and for treating the subject for heart fibrillation, the method comprising:

(a) connecting electrodes to the body of the subject and detecting from the electrodes an analog electrical potential which is proportional to the electrical potential generated by the fibrillating heart;

(b) sampling said analog potential for a selected interval of time to obtain a set of time domain samples;

(c) detecting the power distribution of the fibrillating heart by machine transforming said time domain samples to a frequency domain power spectrum;

(d) detecting the median frequency which bisects the energy of said power spectrum;

(e) treating the fibrillating heart by a therapeutic intervention;

(f) thereafter periodically detecting said median frequency; and

(g) treating the fibrillating heart by applying a means for defibrillation when said detected median frequency rises above a selected frequency.

2. The method of claim 1 wherein said selected frequency is approximately within the range of 7 Hz to 8 Hz.

3. The method of claim 2 wherein the selected frequency is approximately 7.7 Hz.

4. A method for measuring a clinically useful characteristic of the fibrillating heart of a subject which correlates with the approximate elapsed time since the onset of ventricular fibrillation of the heart, and with the metabolic state of the heart therapeutic intervention, the method comprising:

(a) connecting electrodes to the body of the subject and detecting from the electrodes an analog electrical potential which is proportional to the electrical potential generated by the fibrillating heart;

(b) sampling said analog potential for a selected interval of time to obtain a set of time domain samples;

(c) detecting the power distribution of the fibrillating heart by machine transforming said time domain samples to a frequency domain power spectrum;

(d) detecting the median frequency which bisects the energy of said power spectrum; and

(e) measuring the approximate elapsed time since the onset of ventricular fibrillation by comparing said measured median frequency to a model median frequency of a composite model of median frequency as a function of time, said model having been statistically derived from a plurality of prior measurements of median frequency and time elapsed since the onset of fibrillation for a plurality of similar subjects and for an elapsed time exceeding three minutes.

5. A method in accordance with claim 4 wherein said composite model is a mathematical model algorithm statistically derived from said measurements of said similar subjects.

6. A method in accordance with claim 5 wherein said model algorithm is at least two polynomials of at least first order.

7. A method in accordance with claim 6 wherein said model algorithm comprises a sequence of linear approximations of said prior measurements of similar subjects.

8. A method for measuring a clinically useful characteristic of the fibrillating heart of a subject which correlates with the approximate elapsed time since the onset of ventricular fibrillation of the heart, and with the metabolic state of the heart after therapeutic intervention, the method comprising:

(a) connecting electrodes to the body of the subject and detecting from the electrodes an analog electrical potential which s proportional to the electrical potential generated by the fibrillating heart;

(b) sampling said analog potential for a selected interval of time to obtain a set of time domain samples;

(c) detecting the power distribution of the fibrillating heart by machine transforming said time domain samples to a frequency domain power spectrum;

(d) detecting the median frequency which bisects the energy of said power spectrum; and

(e) measuring the approximate elapsed time since the onset of ventricular fibrillation by comparing said measured median frequency to a model median frequency of a composite model of median frequency as a function of tie, said model having been statistically derived from a plurality of prior measurements of median frequency and time elapsed since the onset of fibrillation for a plurality of similar subjects, wherein said composite model is a mathematical model algorithm statistically derived from said measurements of said similar subjects, wherein said model algorithm is at least two polynomials of at least first order and wherein said polynomials are substantially

FM(t)=2.524762t+5.057598 for a measured median frequency slope greater than

and

FM(t)=-1.057669t+17.043929 for a median frequency slope less than or equal to 0.

9. A method for measuring a clinically useful characteristic of the fibrillating heart of a subject which correlates with the approximate elapsed time since the onset of ventricular fibrillation of the heart, and with the metabolic state of the heart after therapeutic intervention, the method comprising:

(a) connecting electrodes to the body of the subject and detecting from the electrodes an analog electrical potential which is proportional to the electrical potential generated by the fibrillating heart;

(b) sampling said analog potential for a selected interval of time to obtain a set of time domain samples;

(c) detecting the power distribution of the fibrillating heart by machine transforming said time domain samples to a frequency domain power spectrum;

(d) detecting the median frequency which bisects the energy of said power spectrum; and

(e) measuring the approximate elapsed time since the onset of ventricular fibrillation by comparing said measured median frequency to a model median frequency of a composite model of median frequency as a function of time, said model having been statistically derived from a plurality of prior measurements of median frequency and time elapsed since the onset of fibrillation for a plurality of similar subjects and wherein said composite model is a mathematical model algorithm having a at least polynomials of at least first order and is statistically derived from said measurements of said similar subjects,

wherein, before the step for measuring the approximate elapsed time since the onset of ventricular fibrillation the median frequency is detected for at lest four selected time intervals and the slope of the median frequency as a function of time is machine computed and compared to the slope of said composite model to determine the time range of the model to utilize for detecting the approximate elapsed time since the onset of fibrillation.