Pediatr Cardiol 13:72-75, 1992
Pediatric Cardiology 9 Springer-VerlagNew York Inc. 1992
G r a d i n g the Severity o f C o n g e s t i v e H e a r t Failure in Infants Robert D. Ross, Robert O. Bollinger, and William W. Pinsky Division of Cardiology, Department of Pediatrics, Children's Hospital of Michigan, Wayne State University School of Medicine, Detroit, Michigan, USA
SUMMARY. To determine which variables most accurately define congestive heart failure (CHF) in infants, 41 patients (median age 2.5 months) were graded by four pediatric cardiologists for the presence and severity of CHF based on the following variables: amount of formula consumed per feeding, feeding time, history of diaphoresis or tachypnea, growth parameters, respiratory and heart rates, respiratory pattern, perfusion, presence of edema, diastolic filling sounds, and hepatomegaly. There were 19 patients graded as having no CHF, nine as mild, seven moderate, and six severe CHF. The most sensitive and specific variables (p < 0.0001) for the presence of CHF were a history of 50/min, an abnormal respiratory pattern, diastolic filling sounds, and hepatomegaly. Moderate to severe CHF was present when patients took 40 min/feed, had an abnormal respiratory pattern with a resting respiratory rate >60/min, and had a diastolic filling sound and moderate hepatomegaly. Severe CHF was accompanied by a heart rate >170/min, decreased perfusion, and severe hepatomegaly. Thus, the grading of the severity of CHF in infants should include an accurate description of these historical and clinical variables. KEY WORDS: Congestive heart failure m Congenital heart disease
The term congestive heart failure (CHF) denotes a pathophysiologic state associated with decreased perfusion such that the rate of blood flow to the tissues does not meet the metabolic requirements of the body. At present, there is no systematic, reproducible method for grading the severity of CHF in infants. In adults, the New York Heart Association (NYHA) classification is widely used to relate the symptoms of CHF to the amount of effort required to provoke them [12]. In the pediatric population, the problem of grading severity of CHF is made difficult by the inability to judge accurately features of the NYHA classification, such as dyspnea on exertion, paroxysmal nocturnal dyspnea, palpitations, neck vein distension, and jugular venous pressure. Therefore, this study was performed to determine which historical variables and examination findings are most closely associated with CHF in infants.
Methods
Subjects The study population was limited to infants to provide a homogeneous group with respect to baseline vital signs. Patients were enlisted for study if they were new to the cardiology department and if their families gave written informed consent according to the guidelines of the Institutional Review board at Children's Hospital of Michigan. To achieve a balance between patients with and without CHF, potential candidates were screened by a pediatric cardiac nurse clinician who enlisted all families of infants who appeared tachypneic, and a random equal number of those without tachypnea. The cardiologists who performed the evaluations were blinded to patient identity and diagnosis. Subjects were excluded if they were acutely febrile to avoid changes in vital signs produced by temperature elevations.
Protocol The infant's primary caretaker was asked the following questions:
Address offprint requests to: Dr. Robert D. Ross, Division of Cardiology, Children's Hospital of Michigan, 3901 Beaubien Blvd., Detroit, MI 48201, USA.
1. How many ounces of formula are consumed per feeding? 2. How long (in minutes) does a feeding take?
Ross et al.: Congestive Heart Failure
73
Table 1. Grading patterns Examiner
Patients quiet for all 4 e x a m i n e r s
Total n
1 2 3 4
0 0 0 0
0 0 0 1
0 0 1 1
0 1 1 2
1 1 1 1
0 1 2 2
1 1 1 2
1 1 2 2
1 2 2 2
2 2 2 2
3 3 2 3
n
16
1
3
1
2
1
1
2
2
1
2
32
Patients upset for 1 or more e x a m i n e r s 1 2 3 4
0* 0* 0* 0
0 0 0* 0
1 2 1 0*
1" 2* 2* 2
2 2 2 1"
2 3 3* 2
2* 3 2 3
2* 2* 3 3
3* 2 2 3
n
1
1
1
1
1
1
1
l
1
9 41
0 = N o C H F ; I = mitd C H F ; 2 = moderate C H F ; 3 = severe C H F (congestive heart failure); *, Patient was u p s e t during examination.
3. My child sweats: (a) not at all, (b) only w h e n it is hot outside, (c) only w h e n upset, (d) with feeding, (e) at rest in cool w e a t h e r (only d or e were considered positive r e s p o n s e s for diaphoresis). 4. M y child has fast or h e a v y breathing: (a) not at all, (b) only w h e n upset, (c) with feeding, (d) at rest (only c or d were considered positive r e s p o n s e s for tachypnea). The patient's birth weight and current weight and height were recorded on a growth chart. Growth failure was defined as a weight less than 5th% for age, or a drop in weight greater than 2 lines on the growth chart [13]. The infants were placed in a quiet examination r o o m where the heart rate and respiratory rate were obtained at rest. E a c h patient was then e x a m i n e d by four board certified or board eligible pediatric cardiologists (from a cohort of nine). Examiners noted on a check sheet: (a) the state of the p a t i e n t - quiet, active, or upset; (b) their respiratory p a t t e r n - - n o r m a l or abnormal (i.e., retractions, grunting, nasal flaring); (c) peripheral skin p e r f u s i o n - - n o r m a l or decreased; (d) peripheral e d e m a - absent or present; (e) d i a p h o r e s i s - - a b s e n t or present; (f) $3 or diastolic r u m b l e - - a b s e n t or present; and (g) h e p a t o m e g a l y - - a b sent or distance (cm) below the right costal margin. Based on this information, each e x a m i n e r was instructed to then assign an overall grade o f the severity o f C H F as follows: 0 = N O C H F , 1 = MILD CHF, 2 = MODERATE CHF, 3 = SEVERE CHF.
Statistics Sensitivity and specificity analyses were. performed using the average grade from the four examinations to determine the presence or absence of C H F in each patient. T h e variables from the history and physical examination were evaluated using the categories: no C H F vs. C H F of any degree; no or mild C H F vs. moderate to severe C H F ; and no, mild, or moderate C H F vs. severe C H F . The cut-off values for the continuous variables were varied to determine which cut-off proved most sensitive and specific. Using the 2 x 2 tables so constructed, a F i s h e r ' s
exact test determined which association were statistically significant similar to the m e t h o d s of Guyatt et al. [6]. Since a large n u m b e r of variables were tested, a stringent alpha level of p < 0.01 was used as a criterion for statistical significance.
Results Thirty-six of the 41 infants enrolled for study were evaluated in the outpatient clinic, and five were examined at the bedside soon after hospital admission. The median age was 2.5 months, with 37 of 41 infants under 5 months of age. After completion of the study, the anatomic diagnoses of the patients were determined to be ventricular septal defects in 17, other congenital heart defects producing left to right shunts in 10, functional heart murmurs in five, mild pulmonary stenosis in five, congestive cardiomyopathy in three, and one anomalous origin of the left coronary artery from the pulmonary artery. Table 1 shows the grade patterns obtained by the cardiologists based on the historical information and their physical findings. Thirty-two of the 41 infants were quiet during the four examinations, while nine were upset for at least one examiner. Of these nine, seven were graded as having CHF, while only 15 of the 32 quiet patients had CHF. Since only three of the 41 infants were upset for more than one of their examinations, no patient was excluded based on their degree of agitation. Overall, there was concordance among the examiners with 100% examiner agreement found on 23 infants (56%). An additional six had 75% examiner agreement, and nine had 50% agreement. Only
74
three patients (7%) had examiners disagreeing by more than one grade of CHF. Averaging the four grades for each subject gave 19 infants without C H F (grade < 0.5), nine with mild C H F (grade of 0.5-1.4), seven with moderate C H F (grade of 1.52.4), and six with severe C H F (grade > 2.4). The variables most sensitive and specific for the presence of C H F were a history of < 3.5 ounces of formula consumed per feeding (the two breastfed infants fed " n o r m a l l y " by history and had no CHF), a respiratory rate > 50/min, an abnormal respiratory pattern, a diastolic filling sound, and hepatomegaly (Table 2A). Additional variables useful for discriminating moderate to severe C H F from mild or no C H F were a history of < 3 ounces/feeding or > 40 min/feeding, a history of tachypnea, an observed respiratory rate (with the patient quiet) > 60/min, and the liver edge > 2.5 cm below the right costal margin (Table 2B). Severe C H F was indicated by a heart rate > 170/rain, decreased peripheral perfusion, and a greater degree of hepatomegaly (Table 2C). N o patient in this series had peripheral edema, and neither diaphoresis nor growth failure were helpful in predicting C H F . While growth failure was quite specific for the presence of C H F , its lack of sensitivity likely is related to the early age of these infants at time of study, before many of those with C H F had a chance to deviate significantly from established growth curves.
Discussion
Defining exactly what constitutes C H F in infants and grading its severity are difficult tasks. In the past, the N Y H A classification has been most commonly utilized to describe functional cardiac status; however, there exists some confusion in the literature as to what each class represents [12]. This is illustrated by two recent articles where the clinical N Y H A classification contained an additional but undefined group labeled class V [8, 9]. In another study, four patients were labeled class I (no limitation) despite the authors' statement that "all patients had some limitation in functional capacity as assessed by N e w York Heart Association classificat i o n " [10]. Even more concerning is the extrapolation of the N Y H A classification to infants and children for evaluation of the severity of C H F . Numerous examples exist of authors describing infants and children as N H Y A class I through IV without a reference or any further definition of these terms [1-5, 7, 11]. In addition to the problem of subjectivity inherent to this system, there is obvious difficulty in reli-
Pediatric Cardiology Vol. 13, No. 2, 1992
Table 2. Predictors of Congestive Heart Failure (CHF)
Variable
Sensitivity
Specificity
p value~
A. For any degree of CHF (average grade ->0.5) History 30 rain/feed 0.55 0.79 NS Diaphoresis 0.50 0.79 NS Tachypnea 0.86 0.63 --* Physical exam Resp. rate >50 0.91 0.84 --*** Heart rate >150 0.55 0.84 NS Growth failure 0.41 1.0 --* Resp. pattern 0.95 0.84 --*** Diaphoresis 0.41 1.0 --* Perfusion 0.31 1.0 --* $3 or rumble 1.0 0.79 --*** Liver edge >1.5 cm 0.95 0.95 --*** B. For moderate to severe CHF (average grade >-1.5) History 40 rain/feed 0.62 0.93 --** Diaphoresis 0.69 0.79 --* Tachypnea 1.0 0.54 --** Physical exam Resp. rate >60 0.85 0.79 --** Heart rate >160 0.54 0.89 --* Growth failure 0.38 0.86 NS Resp. pattern 1.0 0.61 --** Diaphoresis 0.46 0.89 NS Perfusion 0.46 0.96 --* $3 or rumble 1.0 0.54 --** Liver edge >2.5 cm 0.85 0.96 --*** C. For severe CHF (average grade >-2.5) History 40 min/feed 0.83 0.86 --* Diaphoresis 0.83 0.71 NS Tachypnea ! .0 0.43 NS Physical exam Resp. rate >60 0.83 0.66 NS Heart rate >170 0.83 1.0 --*** Growth failure 0.50 0.83 NS Resp. pattern 1.0 0.49 NS Diaphoresis 0.67 0.86 NS Perfusion 0.83 0.94 --** $3 or rumble 1.0 0.43 NS Liver edge >3.0 cm 0.83 0.89 --** Fisher's exact test on 2 x 2 table. *,
Pediatric Cardiology 9 Springer-VerlagNew York Inc. 1992
G r a d i n g the Severity o f C o n g e s t i v e H e a r t Failure in Infants Robert D. Ross, Robert O. Bollinger, and William W. Pinsky Division of Cardiology, Department of Pediatrics, Children's Hospital of Michigan, Wayne State University School of Medicine, Detroit, Michigan, USA
SUMMARY. To determine which variables most accurately define congestive heart failure (CHF) in infants, 41 patients (median age 2.5 months) were graded by four pediatric cardiologists for the presence and severity of CHF based on the following variables: amount of formula consumed per feeding, feeding time, history of diaphoresis or tachypnea, growth parameters, respiratory and heart rates, respiratory pattern, perfusion, presence of edema, diastolic filling sounds, and hepatomegaly. There were 19 patients graded as having no CHF, nine as mild, seven moderate, and six severe CHF. The most sensitive and specific variables (p < 0.0001) for the presence of CHF were a history of 50/min, an abnormal respiratory pattern, diastolic filling sounds, and hepatomegaly. Moderate to severe CHF was present when patients took 40 min/feed, had an abnormal respiratory pattern with a resting respiratory rate >60/min, and had a diastolic filling sound and moderate hepatomegaly. Severe CHF was accompanied by a heart rate >170/min, decreased perfusion, and severe hepatomegaly. Thus, the grading of the severity of CHF in infants should include an accurate description of these historical and clinical variables. KEY WORDS: Congestive heart failure m Congenital heart disease
The term congestive heart failure (CHF) denotes a pathophysiologic state associated with decreased perfusion such that the rate of blood flow to the tissues does not meet the metabolic requirements of the body. At present, there is no systematic, reproducible method for grading the severity of CHF in infants. In adults, the New York Heart Association (NYHA) classification is widely used to relate the symptoms of CHF to the amount of effort required to provoke them [12]. In the pediatric population, the problem of grading severity of CHF is made difficult by the inability to judge accurately features of the NYHA classification, such as dyspnea on exertion, paroxysmal nocturnal dyspnea, palpitations, neck vein distension, and jugular venous pressure. Therefore, this study was performed to determine which historical variables and examination findings are most closely associated with CHF in infants.
Methods
Subjects The study population was limited to infants to provide a homogeneous group with respect to baseline vital signs. Patients were enlisted for study if they were new to the cardiology department and if their families gave written informed consent according to the guidelines of the Institutional Review board at Children's Hospital of Michigan. To achieve a balance between patients with and without CHF, potential candidates were screened by a pediatric cardiac nurse clinician who enlisted all families of infants who appeared tachypneic, and a random equal number of those without tachypnea. The cardiologists who performed the evaluations were blinded to patient identity and diagnosis. Subjects were excluded if they were acutely febrile to avoid changes in vital signs produced by temperature elevations.
Protocol The infant's primary caretaker was asked the following questions:
Address offprint requests to: Dr. Robert D. Ross, Division of Cardiology, Children's Hospital of Michigan, 3901 Beaubien Blvd., Detroit, MI 48201, USA.
1. How many ounces of formula are consumed per feeding? 2. How long (in minutes) does a feeding take?
Ross et al.: Congestive Heart Failure
73
Table 1. Grading patterns Examiner
Patients quiet for all 4 e x a m i n e r s
Total n
1 2 3 4
0 0 0 0
0 0 0 1
0 0 1 1
0 1 1 2
1 1 1 1
0 1 2 2
1 1 1 2
1 1 2 2
1 2 2 2
2 2 2 2
3 3 2 3
n
16
1
3
1
2
1
1
2
2
1
2
32
Patients upset for 1 or more e x a m i n e r s 1 2 3 4
0* 0* 0* 0
0 0 0* 0
1 2 1 0*
1" 2* 2* 2
2 2 2 1"
2 3 3* 2
2* 3 2 3
2* 2* 3 3
3* 2 2 3
n
1
1
1
1
1
1
1
l
1
9 41
0 = N o C H F ; I = mitd C H F ; 2 = moderate C H F ; 3 = severe C H F (congestive heart failure); *, Patient was u p s e t during examination.
3. My child sweats: (a) not at all, (b) only w h e n it is hot outside, (c) only w h e n upset, (d) with feeding, (e) at rest in cool w e a t h e r (only d or e were considered positive r e s p o n s e s for diaphoresis). 4. M y child has fast or h e a v y breathing: (a) not at all, (b) only w h e n upset, (c) with feeding, (d) at rest (only c or d were considered positive r e s p o n s e s for tachypnea). The patient's birth weight and current weight and height were recorded on a growth chart. Growth failure was defined as a weight less than 5th% for age, or a drop in weight greater than 2 lines on the growth chart [13]. The infants were placed in a quiet examination r o o m where the heart rate and respiratory rate were obtained at rest. E a c h patient was then e x a m i n e d by four board certified or board eligible pediatric cardiologists (from a cohort of nine). Examiners noted on a check sheet: (a) the state of the p a t i e n t - quiet, active, or upset; (b) their respiratory p a t t e r n - - n o r m a l or abnormal (i.e., retractions, grunting, nasal flaring); (c) peripheral skin p e r f u s i o n - - n o r m a l or decreased; (d) peripheral e d e m a - absent or present; (e) d i a p h o r e s i s - - a b s e n t or present; (f) $3 or diastolic r u m b l e - - a b s e n t or present; and (g) h e p a t o m e g a l y - - a b sent or distance (cm) below the right costal margin. Based on this information, each e x a m i n e r was instructed to then assign an overall grade o f the severity o f C H F as follows: 0 = N O C H F , 1 = MILD CHF, 2 = MODERATE CHF, 3 = SEVERE CHF.
Statistics Sensitivity and specificity analyses were. performed using the average grade from the four examinations to determine the presence or absence of C H F in each patient. T h e variables from the history and physical examination were evaluated using the categories: no C H F vs. C H F of any degree; no or mild C H F vs. moderate to severe C H F ; and no, mild, or moderate C H F vs. severe C H F . The cut-off values for the continuous variables were varied to determine which cut-off proved most sensitive and specific. Using the 2 x 2 tables so constructed, a F i s h e r ' s
exact test determined which association were statistically significant similar to the m e t h o d s of Guyatt et al. [6]. Since a large n u m b e r of variables were tested, a stringent alpha level of p < 0.01 was used as a criterion for statistical significance.
Results Thirty-six of the 41 infants enrolled for study were evaluated in the outpatient clinic, and five were examined at the bedside soon after hospital admission. The median age was 2.5 months, with 37 of 41 infants under 5 months of age. After completion of the study, the anatomic diagnoses of the patients were determined to be ventricular septal defects in 17, other congenital heart defects producing left to right shunts in 10, functional heart murmurs in five, mild pulmonary stenosis in five, congestive cardiomyopathy in three, and one anomalous origin of the left coronary artery from the pulmonary artery. Table 1 shows the grade patterns obtained by the cardiologists based on the historical information and their physical findings. Thirty-two of the 41 infants were quiet during the four examinations, while nine were upset for at least one examiner. Of these nine, seven were graded as having CHF, while only 15 of the 32 quiet patients had CHF. Since only three of the 41 infants were upset for more than one of their examinations, no patient was excluded based on their degree of agitation. Overall, there was concordance among the examiners with 100% examiner agreement found on 23 infants (56%). An additional six had 75% examiner agreement, and nine had 50% agreement. Only
74
three patients (7%) had examiners disagreeing by more than one grade of CHF. Averaging the four grades for each subject gave 19 infants without C H F (grade < 0.5), nine with mild C H F (grade of 0.5-1.4), seven with moderate C H F (grade of 1.52.4), and six with severe C H F (grade > 2.4). The variables most sensitive and specific for the presence of C H F were a history of < 3.5 ounces of formula consumed per feeding (the two breastfed infants fed " n o r m a l l y " by history and had no CHF), a respiratory rate > 50/min, an abnormal respiratory pattern, a diastolic filling sound, and hepatomegaly (Table 2A). Additional variables useful for discriminating moderate to severe C H F from mild or no C H F were a history of < 3 ounces/feeding or > 40 min/feeding, a history of tachypnea, an observed respiratory rate (with the patient quiet) > 60/min, and the liver edge > 2.5 cm below the right costal margin (Table 2B). Severe C H F was indicated by a heart rate > 170/rain, decreased peripheral perfusion, and a greater degree of hepatomegaly (Table 2C). N o patient in this series had peripheral edema, and neither diaphoresis nor growth failure were helpful in predicting C H F . While growth failure was quite specific for the presence of C H F , its lack of sensitivity likely is related to the early age of these infants at time of study, before many of those with C H F had a chance to deviate significantly from established growth curves.
Discussion
Defining exactly what constitutes C H F in infants and grading its severity are difficult tasks. In the past, the N Y H A classification has been most commonly utilized to describe functional cardiac status; however, there exists some confusion in the literature as to what each class represents [12]. This is illustrated by two recent articles where the clinical N Y H A classification contained an additional but undefined group labeled class V [8, 9]. In another study, four patients were labeled class I (no limitation) despite the authors' statement that "all patients had some limitation in functional capacity as assessed by N e w York Heart Association classificat i o n " [10]. Even more concerning is the extrapolation of the N Y H A classification to infants and children for evaluation of the severity of C H F . Numerous examples exist of authors describing infants and children as N H Y A class I through IV without a reference or any further definition of these terms [1-5, 7, 11]. In addition to the problem of subjectivity inherent to this system, there is obvious difficulty in reli-
Pediatric Cardiology Vol. 13, No. 2, 1992
Table 2. Predictors of Congestive Heart Failure (CHF)
Variable
Sensitivity
Specificity
p value~
A. For any degree of CHF (average grade ->0.5) History 30 rain/feed 0.55 0.79 NS Diaphoresis 0.50 0.79 NS Tachypnea 0.86 0.63 --* Physical exam Resp. rate >50 0.91 0.84 --*** Heart rate >150 0.55 0.84 NS Growth failure 0.41 1.0 --* Resp. pattern 0.95 0.84 --*** Diaphoresis 0.41 1.0 --* Perfusion 0.31 1.0 --* $3 or rumble 1.0 0.79 --*** Liver edge >1.5 cm 0.95 0.95 --*** B. For moderate to severe CHF (average grade >-1.5) History 40 rain/feed 0.62 0.93 --** Diaphoresis 0.69 0.79 --* Tachypnea 1.0 0.54 --** Physical exam Resp. rate >60 0.85 0.79 --** Heart rate >160 0.54 0.89 --* Growth failure 0.38 0.86 NS Resp. pattern 1.0 0.61 --** Diaphoresis 0.46 0.89 NS Perfusion 0.46 0.96 --* $3 or rumble 1.0 0.54 --** Liver edge >2.5 cm 0.85 0.96 --*** C. For severe CHF (average grade >-2.5) History 40 min/feed 0.83 0.86 --* Diaphoresis 0.83 0.71 NS Tachypnea ! .0 0.43 NS Physical exam Resp. rate >60 0.83 0.66 NS Heart rate >170 0.83 1.0 --*** Growth failure 0.50 0.83 NS Resp. pattern 1.0 0.49 NS Diaphoresis 0.67 0.86 NS Perfusion 0.83 0.94 --** $3 or rumble 1.0 0.43 NS Liver edge >3.0 cm 0.83 0.89 --** Fisher's exact test on 2 x 2 table. *,
