6 0 208 KB
See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/223958463
The Ross Classification for Heart Failure in Children After 25 Years: A Review and an Age-Stratified Revision Article in Pediatric Cardiology · April 2012 DOI: 10.1007/s00246-012-0306-8 · Source: PubMed
CITATIONS
READS
41
8,358
1 author: Robert Ross Detroit Medical Center 62 PUBLICATIONS 1,643 CITATIONS SEE PROFILE
All content following this page was uploaded by Robert Ross on 25 November 2014.
The user has requested enhancement of the downloaded file.
Pediatr Cardiol DOI 10.1007/s00246-012-0306-8
ORIGINAL ARTICLE
The Ross Classification for Heart Failure in Children After 25 Years: A Review and an Age-Stratified Revision Robert D. Ross
Received: 10 January 2012 / Accepted: 14 March 2012 Ó Springer Science+Business Media, LLC 2012
Abstract Accurate grading of the presence and severity of heart failure (HF) signs and symptoms in infants and children remains challenging. It has been 25 years since the Ross classification was first used for this purpose. Since then, several modifications of the system have been used and others proposed. New evidence has shown that in addition to signs and symptoms, data from echocardiography, exercise testing, and biomarkers such as N-terminal pro-brain natriuretic peptide (NT-proBNP) all are useful in stratifying outcomes for children with HF. It also is apparent that grading of signs and symptoms in children is dependent on age because infants manifest HF differently than toddlers and older children. This review culminates in a proposed new age-based Ross classification for HF in children that incorporates the most useful data from the last two decades. Testing of this new system will be important to determine whether an age-stratified scoring system can unify the way communication of HF severity and research on HF in children is performed in the future. Keywords Age-based Ross classification Heart failure Ross classification
The grading of heart failure (HF) signs and symptoms in infants and children remains challenging. Ideally, a system for doing this would be accurate, reproducible, correlated closely with disease severity and outcome, and fluid to
R. D. Ross (&) Division of Pediatric Cardiology, Carmen and Ann Adams Department of Pediatrics, Children’s Hospital of Michigan, Wayne State University School of Medicine, 3901 Beaubien Blvd., Detroit, MI 48201, USA e-mail: [email protected]
reflect changes in symptoms over time and with therapy. The classification also should predict risk from disease so that management can be tailored to HF class. Children classified as ‘‘no risk’’ would not require treatment. ‘‘Mild risk’’ might be managed by closer observation, early intervention, or even prophylaxis. ‘‘Moderate risk’’ would engender more intensive treatment, and ‘‘severe risk’’ would require maximal therapy and perhaps transplantation referral. Until 1987, the only system available for grading HF in children was the New York Heart Association (NYHA) classification. However, this system was based on limitations to physical activity for adults, which did not translate well for use with children, particularly infants. Therefore, we developed a symptom-based classification using more age-appropriate variables (Table 1) and demonstrated that plasma norepinephrine correlated in a stepwise fashion with this new Ross HF classification from grades I to IV [16]. This was significant in that it mirrored norepinephrine changes in adults with HF that correlated closely with mortality [3]. Subsequently, Wu et al. [25] confirmed the catecholamine changes with Ross class and further showed that progressive beta receptor downregulation occurred with each progressive Ross class. More recently, Fernandes et al. [7] evaluated children with idiopathic dilated cardiomyopathy. These authors found that both the presence and severity of mitral regurgitation (MR) increased with the Ross class and that the presence of MR stratified children to a significantly worse outcome of either death or transplantation, with a hazard ratio of 1.9 (p \ 0.01). Progression of MR severity increased these risks significantly. Over time, use of the Ross classification system has increased, but the system is associated with problems. Assignment of classes I and IV tends to be straightforward,
123
Pediatr Cardiol Table 1 Original Ross classification [16] I: No limitations or symptoms II: Mild tachypnea or diaphoresis with feedings in infants, dyspnea at exertion in older children; no growth failure III: Marked tachypnea or diaphoresis with feedings or exertion and prolonged feeding times with growth failure from CHF IV: Symptomatic at rest with tachypnea, retractions, grunting, or diaphoresis
with either no symptoms (class I) or symptoms with the patient at rest (class IV). However, classes II and III are more subjective and can overlap. Also, few class IV patients are found in children who may have a greater ability to compensate for HF early on. This leads to many studies combining classes III and IV to have adequate numbers for data analysis. In addition, growth failure may be the only manifestation of HF in young children, which makes it difficult to determine whether that puts them in Ross class III or a lower class in the absence of respiratory or other symptoms listed. Also confusing is the definition of growth failure which has differed over time with various authors, and growth failure may be related to noncardiac conditions. To evaluate the variables that have most accurately defined HF, we studied 41 infants and used the blinded average grade from four pediatric cardiologists to compare signs and symptoms. Based on the most sensitive and specific variables, a scoring system for grading HF in infants was derived (Table 2) [17]. Interestingly, neither diaphoresis nor growth failure proved to be significant. This is likely due to the frequent sweating that normal infants exhibit and the young median age of 2.5 months in
the study, which may have been too early for growth failure from HF to have become manifest. Several authors have modified the aforementioned scoring system to expand its use to older children. The first such attempt was by Reithmann et al. [14] in their study on adenylyl cyclase in severe HF. They added basal heart and respiratory rates for children by age, namely, for infants and children 1–6 years old, children 7–10 years old, and children 11–14 years old. This was a useful modification, but these authors also reinstituted diaphoresis and added cyanosis and a precordial thrill, neither of which are typically associated with HF. In 2002, Laer et al. [10] modified this further, limiting their new version to six variables (Table 3). A potential downside of this simplified version is that half of the six categories relate to the work of breathing, which may undervalue the other manifestations of congestive heart failure (CHF) in children. Nevertheless, many subsequent studies have used this Laer-modified Ross scoring system to study HF. At about the same time, a new system called the New York University Pediatric Heart Failure Index (PHFI) was proposed by Connelly et al. [4]. This 30-point scale uses many of the signs and symptoms in the Ross classification but adds points for medications used to treat HF and also for a single-ventricle physiology. Although sicker patients generally do require more medications, scoring in this way may be problematic. If a child with severe symptoms of HF is treated and the symptoms improve, the change in score
Table 3 Modified Ross score [10] 0
?1
?2
Diaphoresis
Head only
Head and body at exertion
Head and body at rest
Tachypnea
Rare
Several times
Frequent
Retractions
Dyspnea
History Table 2 Ross scoring system for heart failure in infants [17] Score 0
1
2
Breathing
Feeding history Volume consumed per feeding (oz)
[3.5
2.5–3.5
\2.5
Time taken per feeding (min)
\40
[40
–
Physical exam Respiratory rate (n/min)
\50
50–60
[60
Heart rate (n/min)
\160
160–170
[170
Respiratory pattern
Normal
Abnormal
–
Peripheral perfusion
Normal
Decreased
–
S3 or diastolic rumble
Absent
Present
–
Liver edge from right costal margin \2 (cm)
2–3
[3
Total score: 0–2 (no CHF), 3–6 (mild CHF), 7–9 (moderate CHF), 10–12 (severe CHF) CHF congestive heart failure
123
Physical examination Normal
Age (years) Respiratory rate (breaths/min) (years) 0–1 \50 50–60
[60
1–6
\35
35–45
[45
7–10
\25
25–35
[35
11–14
\18
18–28
[28
Heart rate (beats/min) (years) 0–1
\160
160–170
[170
1–6
\105
105–115
[l15
7–10
\90
90–100
[100
11–14
\80
80–90
[90
\2
2–3
[3
Hepatomegaly size (cm)
Pediatr Cardiol
will be blunted by the points received for these medications and thus may not reflect the improvement [15]. Regarding single-ventricle patients, while most have reduced exercise capacity, some children with Fontan palliation have no symptoms, experience normal aerobic capacity for age and body size, and thus do not deserve higher HF scores. In 2006, Tissieres et al. [23] compared the NYHA classification, the Laer-modified Ross classification, and the PHFI using 20 children with HF from rheumatic heart disease. Although all three systems correlated with the cardiothoracic index on chest X-ray, the PHFI faired better on left ventricle (LV) mass, end-systolic wall stress, left atrium/aortic ratio, and N-terminal pro-brain natriuretic peptide (NT-proBNP). Recently, a great deal of interest in both the adult and pediatric HF literature has focused on the natriuretic peptides in HF. From myocardial cells, BNP is released into the bloodstream in response to various stressors on the heart including LV volume and pressure overload. This correlates well with symptoms of HF in adults and children and can differentiate cardiac from pulmonary causes of respiratory distress [6]. As the N-terminal fragment of the prohormone BNP, NT-proBNP is a good marker of clinical severity and worsening systolic function in children with HF [13] and has a longer half-life than BNP. Sugimoto et al. [22] found very sensitive and specific cutoff points of NT-proBNP for Ross classes I to IV that had area-underthe receiver operating curves of 0.96 to 0.99. There was a dichotomy of values, with lower numbers for children older than 3 years than for children younger than 3 years. For distinguishing each class independently, NT-proBNP was better than BNP itself. Multiple other studies have confirmed the usefulness of NT-proBNP and BNP as correlates of HF symptoms in children, as markers of systolic dysfunction, and importantly, as predictors of the need for mechanical circulatory support, heart transplantation, and death [1, 18, 24]. This has held true for HF from cardiomyopathy and from congenital heart disease such as single ventricle with failing Fontan palliation [11, 20]. The trend of NT-proBNP over time in individual patients is most useful for predicting outcomes [24]. Another factor found to be useful for predicting outcomes in HF is exercise capacity. In adults, a peak exercise oxygen consumption of less than 14 ml/kg/min is an independent predictor of mortality and a criterion for listing a patient to receive a heart transplant. However, this cutoff may be less sensitive in the current era of improved medical management for HF [2]. In addition, children have different oxygen consumptions as they grow such that the absolute number of 14 ml/kg/min is not a sensitive marker of the need for transplantation [5]. Giardini et al. [8] performed exercise stress tests on 82 children with dilated
cardiomyopathy and found that in a 4-year follow-up period, the percentage of peak oxygen consumption (based on age and sex) stratified outcomes accurately. Using 62 % of predicted normal as a cutoff, survival curves indicated a significantly higher rate of death or urgent listing for transplantation of 50.6 (for those B62 %) versus 4.4 % (for those [62 %) at 24 months, with a hazard ratio of 10.8. Growing evidence also indicates that poor systolic function bodes ill in terms of long-term outcomes for children as it does for adults. In both dilated cardiomyopathy and HF from congenital heart disease, low ejection fractions predict death or the need for transplantation [9, 12]. It is clear that with all this recent data on factors predictive of outcomes in children with HF that a revision in how we grade symptom severity is required. It also is apparent that an age stratification is required to encompass the changes in signs and symptoms that children manifest from infancy to late childhood. A classification system should include the biomarkers, echo parameters of systolic function and mitral or systemic atrioventricular valve (AV) insufficiency, and reflect exercise limitations reflected by feeding and growth in infants and exercise capacity indicated by percentage of predicted maximal oxygen uptake (VO2) in older children. Therefore, I propose an age-based Ross classification using the original variables that proved to be sensitive and specific and adding the new evidence-based data. Table 4 depicts this revised system. The age ranges of 0–3 months, 4–12 months, 1–3 years, 4–8 years, and 9–18 years were chosen because the variables in the classification are generally stable during these periods but vary between them. Each age range has 10 variables with scores of 0, 1, or 2 possible for a range of 0 to 20. The scoring system can be used as a continuous data set for comparison with outcomes, or it can be categorized by points assessed as Ross classes I (0–5), II (6–10), III (11–15) and IV (16–20). For all children, hepatomegaly is measured as the distance below the right costal margin with abdominal situs solitus or from the left costal margin for situs inversus. The ejection fraction generally is obtained from echocardiography but can be derived from MRI or other imaging methods for single ventricles or systemic right ventricles. Systemic AV insufficiency refers to the mitral valve for systemic left ventricles and to the systemic AV valve for single ventricles or systemic right ventricles. Each age range has unique aspects that require comment. All heart rate and respiratory rates should be recorded with the infant or child in the basal state without crying or undo agitation, and the cutoff points have been selected based on normals for these age ranges [19].
123
Pediatr Cardiol Table 4 Age-based Ross classification for heart failure in children 0
1
2
Oz/feeding
[3.5
2.5–3.5
\2.5
Time for feeding (min)
\20
20–40
[40
Breathing
Nl
Tachypnea
Retractions
RR/min
\50
50–60
HR/min
\160
Perfusion Hepatomegaly (cm)
Table 4 continued 0
1
2
N/V
Nl
Intermittent
Frequent
Breathing
Nl
Tachypnea
Retractions
RR/min
\20
20–30
[30
HR/min
\90
90–100
[100
[60
Perfusion
Nl
Reduced
Shocky
160–170
[170
Hepatomegaly (cm)
\2
2–3
[3
Nl \2
Reduced 2–3
Shocky [3
NT-proBNP (pg/ml)
\300
300–1,500
[1,500
EF%
[50
30–50
\30
NT-proBNP (pg/ml)
\450 ([4 days)
450–1,700
[1,700
Max %VO2
[80
60–80
\60
EF%
[50
30–50
\30
AV insufficiency
None
Mild
Moderate/ severe
AV insufficiency
None
Mild
Moderate/ severe
Feeding
Nl
Decreased
Gavaged
Wt% Breathing
Nl Nl
C1 Curve Tachypnea
C2 Curve Retractions
RR/min
\40
40–50
[50
HR/min
\12
120–130
[130
Perfusion
Nl
Reduced
Shocky
Hepatomegaly (cm)
\2
2–3
[3
NT-proBNP (pg/ml)
\450
450–1,700
[1,700
EF%
[50
30–50
\30
AV insufficiency
None
Mild
Moderate/ severe
Feeding
Nl
Decreased
Gavaged
Growth Breathing
Nl Nl
Weight loss Tachypnea
Cachexia Retractions
RR/min
\30
30–40
[40
HR/min
\110
110–120
[120
Perfusion
Nl
Reduced
Shocky
Hepatomegaly (cm)
\2
2–3
[3
0–3 Months
9–18 Years
4–12 Months
Oz ounce, Nl normal, Wt% fall-off on weight curve %, RR respiratory rate, HR heart rate, NT-proBNP N-terminal pro-brain natriuretic peptide, EF ejection fraction, AV systemic atrioventricular valve, N/ V nausea/vomiting; Max %VO2 % of predicted maximal oxygen uptake for age and sex
Age 0–3 Months The volume of formula per feeding is for bottle-fed babies. For breastfed infants, the volume taken has to be rated subjectively as normal, decreased, or gavage supplemented. Normally, NT-proBNP is elevated in newborns, so this measurement should be obtained after 4 days of life.
1–3 Years
NT-proBNP (pg/ml)
\450
450–1,700
[1,700
EF%
[50
30–50
\30
AV insufficiency
None
Mild
Moderate/ severe
4–8 Years N/V
None
Intermittent
Frequent
Growth
Nl
Weight loss
Cachexia
Breathing RR/min
Nl \25
Tachypnea 25–35
Retractions [35
HR/min
\100
90–100
[100
Perfusion
Nl
Reduced
Shocky
Hepatomegaly (cm)
\2
2–3
[3
NT-proBNP (pg/ml)
\300
300–1,500
[1,500
EF%
[50
30–50
\30
AV insufficiency
None
Mild
Moderate/ severe
123
Age 4–12 Months Feeding is qualitatively graded because diets vary in this age range and specific volumes of formula are not applicable. The time of feeding is replaced by growth, as depicted on the growth curve. A fall-off in growth, defined as a decrease of C1 weight curve percentile (i.e., from the 50 to the 25 %) earns 1 point, whereas a fall-off of C2 percentile curves (i.e., from 50 to 10 %) earns 2 points. If no previous weights are available, then 1 point is awarded for a weight percentage of C1 curve below the current height percentile and 2 points for C2 curve percentiles for weight below that for height because increases in body length typically are preserved in HF, whereas weight gain is not.
Age 1–3 Years Because the time of early rapid growth has passed by these ages, the pattern on the growth curves has been changed to recent weight loss and cachexia for respectively 1 and 2 points.
Pediatr Cardiol
Age 4–8 Years As children age, their gastrointestinal symptoms from HF change to reports of nausea or vomiting, so scores are 1 for these symptoms intermittently and 2 for frequent nausea or vomiting. The cutoff values for NT-proBNP have been adjusted down for the lower values found after the age of 3 years.
differences in symptoms. Standardizing our approach for future research and communication using one system that incorporates all the significant features of HF culled from the literature is a big step toward an evidence-based approach to studying and treating childhood HF in the future.
References Age 9–18 Years At this age, most children can perform a maximal stress test, so this has replaced growth failure as a more objective measure of heart failure decompensation.
BNP Although NT-proBNP has a longer half-life than BNP and correlates better with symptom class, retrospective studies may have access only to BNP levels. If so, then BNP may replace NT-proBNP in the grid for all ages beyond 4 days, with points of 0 for less than 30 pg/ml, 1 for 30–140 pg/ml, and 2 for more than 140 pg/ml. Because these peptides also may be elevated from renal dysfunction, children with advanced renal insufficiency may have a somewhat inflated score for this component [21].
Class IV Finally, some children will be so sick from HF that they will require intravenous medications, circulatory support such as extracorporeal membrane oxygenation (ECMO) or a ventricular assist device, and mechanical ventilation. These interventions make the children unfit for the use of this grading system and should put them by definition into class IV.
Future Challenges The next challenge is to test this age-based Ross classification using large numbers of infants and children with and without overt HF to determine whether it accurately predicts outcomes for each age range. Although it does entail more information gathering than the original Ross system, most patients with HF get tested for these variables in addition to the readily obtained historical and physical examination findings. Use of this age-based stratification should allow for more accurate grading, eliminating the variance previously encountered based on age-related
1. Aurbach SR, Richmond ME, Lamour JM et al (2010) BNP levels predict outcome in pediatric heart failure patients: post hoc analysis of the pediatric carvedilol trial. Circ Heart Fail 3:606–611 2. Butler J, Khadim G, Paul KM et al (2004) Selection of patients for heart transplantation in the current era of heart failure therapy. J Am Coll Cardiol 43:787–793 3. Cohn JN, Levine B, Olivari T et al (1984) Plasma norepinephrine as a guide to prognosis in patients with chronic congestive heart failure. N Engl J Med 311:819–823 4. Connelly D, Rutkowski M, Auslender M, Artman M (2001) The New York University pediatric heart failure index: a new method of quantifying chronic heart failure severity in children. J Pediatr 138:644–648 5. Das BB, Taylor AL, Boucek MM, Wolfe RW, Yetman AT (2006) Exercise capacity in pediatric heart transplant candidates: Is there any role for the 14 ml/kg/min guideline? Pediatr Cardiol 27:226–229 6. Evim MS, Ucar B, Kilic Z, Colak O (2010) The value of serum N-terminal pro-brain natriuretic peptide levels in the differential diagnosis and follow-up of congestive cardiac failure and respiratory distress due to pulmonary aetiologies in infants and children. Cardiol Young 20:495–504 7. Fernandes FP, Manlhiot C, McCrindle BW, Mertens L, Kantor PF, Friedberg MK (2011) Usefulness of mitral regurgitation as a marker of increased risk for death or cardiac transplantation in idiopathic dilated cardiomyopathy in children. Am J Cardiol 107:1517–1521 8. Giardini A, Fenton M, Andrews RE, Derrick G, Burch M (2011) Peak oxygen uptake correlates with survival without clinical deterioration in ambulatory children with dilated cardiomyopathy. Circulation 124:1713–1718 9. Guimaraes GV, d’Avila VM, Camargo PR, Moreira LFP, Luces JRL, Bocchi EA (2008) Prognostic value of cardiopulmonary exercise testing in children with heart failure secondary to idiopathic dilated cardiomyopathy in a non-B-blocker setting. Eur J Heart Fail 10:560–565 10. Laer S, Mir TS, Behn F et al (2002) Carvedilol therapy in pediatric patients with congestive heart failure: a study investigating clinical and pharmacokinetic parameters. Am Heart J 143:916–922 11. Lechner E, Gitter R, Mair R et al (2008) Aminoterminal brain natriuretic peptide levels in children and adolescents after Fontan operation correlate with congestive heart failure. Pediatr Cardiol 29:901–905 12. Patel MS, Berg AM, Vincent RN, Mahle WT (2010) Serum parameters and echocardiographic predictors of death or need for transplant in newborns, children, and young adults with heart failure. Am J Cardiol 105:1798–1801 13. Ratnasamy C, Kinnamon DD, Lipshultz SE, Rusconi P (2008) Associations between neurohormonal and inflammatory activation and heart failure in children. Am Heart J 155:527–533 14. Reithmann C, Reber D, Kozlik-Feldmann R et al (1997) A postreceptor defect of adenylyl cyclase in severely failing
123
Pediatr Cardiol
15. 16.
17. 18.
19. 20.
myocardium from children with congenital heart disease. Eur J Pharm 330:79–86 Ross RD (2001) Grading the graders of congestive heart failure in children. J Pediatr 138:618–620 Ross RD, Daniels SR, Schwartz DC, Hannon DW, Shukla R, Kaplan S (1987) Plasma levels of norepinephrine in infants and children with congestive heart failure. Am J Cardiol 59:911–914 Ross RD, Bollinger RO, Pinsky WW (1992) Grading the severity of congestive heart failure in infants. Pediatr Cardiol 13:72–75 Rusconi PG, Ludwig DA, Ratnasamy C et al (2010) Serial measurements of serum NT-proBNP as markers of left ventricular function and remodeling in children with heart failure. Am Heart J 160:776–783 Saunders M, Gorelick MH (2011) Nelson textbook of pediatrics, 19th edn. Saunders Elsevier, Philadelphia, p 280 Shah A, Feraco AM, Harmon C, Tacy T, Fineman JR, Bernstein HS (2009) Usefulness of various plasma biomarkers for diagnosis of heart failure in children with single ventricle physiology. Am J Cardiol 104:1280–1284
123 View publication stats
21. Srisawasdi P, Vanavanan S, Charoenpanichkit C, Kroll MH (2010) The effect of renal dysfunction on BNP, NT-proBNP, and their ratio. Am J Clin Pathol 133:14–23 22. Sugimoto M, Manabe H, Nakau K et al (2010) The role of N-terminal pro-B type natriuretic peptide in the diagnosis of congestive heart failure in children. Circ J 74:998–1005 23. Tissieres P, Aggouon Y, Da Cruz E et al (2006) Comparison of heart failure classifications in children undergoing valvular surgery. J Pediatr 149:210–215 24. Wong DTH, George K, Wilson J et al (2011) Effectiveness of serial increases in amino-terminal pro-B-type natriuretic peptide levels to indicate the need for mechanical circulatory support in children with acute decompensated heart failure. Am J Cardiol 107:573–578 25. Wu JR, Chang HR, Huang TY, Chiang CH, Chen SS (1996) Reduction in lymphocyte B-adrenergic receptor density in infants and children with heart failure secondary to congenital heart disease. Am J Cardiol 77:170–174