An Ambiguity in Child Body Surface Area Measurement

Featured Application

Body Surface Area is an important parameter in medicine, with numerous potential uses. While our work does not reveal any new applications of this parameter, or a new formula to calculate it, it sums up all existing formulae for determining BSA, compares them, shows the risks that are a result of applying the formulae and encourages further discussions and explorations of the topic, some of them already in the process of preparing for publication.

Abstract

Body Surface Area (BSA) and Total Body Surface Area (TBSA) are important indicators used in different domains of medicine, playing a crucial role in oncology, toxicology, and even transplantology or cardiology. Their main (although not sole) use in medicine is to dose medications according to the patient’s BSA/TBSA. Other fields where BSA is commonly employed include the treatment of burns—the fluid doses are according to Parkland’s rule, which is based on BSA estimation. Thus, a proper estimation of BSA value directly influences the patient’s chances of survival—one can easily imagine the consequences of not administering enough fluids in the critical phase of a burned patient’s management. In medical practice, even small deviations in the estimated BSA value may have a significant impact on the patient’s treatment process and, in extreme cases, lead to their death. This problem is particularly important in the treatment of children and adolescents. The aim of our article is to present the most popular formulae used to estimate the BSA value for children in the case of minors and to discuss discrepancies between them. In the case of a 4-week-old neonate, the smallest difference in maximum BSA calculation discrepancies amounts up to 0.22 m², which corresponds to roughly 80% of the typical BSA value for this age group, while for 12-year-old patients this parameter is 0.15 m², equalling about 11% of the standard BSA value for this age category. The maximum deviation between the patterns reaches 0.28 m² for 4-week-olds and over 0.75 m² for 12-year-old children. These notable discrepancies in BSA calculations highlight the necessity for more precise and dependable methods, particularly when dealing with paediatric patients.

1. Introduction

The therapeutic success of nearly all medical treatments depends on accurate drug dosing. In clinical practice, dosing is often based on measurable patient parameters such as body weight or body mass index (BMI). However, these indices are strongly influenced by physiological and pathological factors, including age, hydration status, and body composition, which may lead to inappropriate dose adjustments and compromised therapeutic outcomes [1,2]. For this reason, body surface area (BSA) has become a widely used parameter in medical diagnostics and therapy optimization, offering a more physiologically relevant measure of metabolic mass than body weight alone [3]. BSA is currently applied in numerous clinical areas, including the dosing of chemotherapeutic and antiviral drugs, assessment of renal and cardiac function, and calculation of burn surface area. However, despite its widespread use, there is still no universally accepted method for BSA estimation. Over the past century, dozens of empirical formulas have been proposed—beginning with Meeh’s weight-based equation in 1879 and followed by the height–weight model of Du Bois and Du Bois in 1916 [4]. Subsequent research introduced further refinements and population-specific adaptations, resulting in a large number of formulas that differ in structure, underlying datasets, and regression assumptions. As a result, the same anthropometric data can yield markedly different BSA values depending on the formula used, which may translate into clinically meaningful variations in drug dosing and other diagnostic or therapeutic decisions [5]. Errors in calculating BSA values can result in substantial dose errors that affect the quality of the patient’s treatment [6,7,8,9].

The recent literature continues to address these concerns. Flint, Bronson, Joe M. Das, and Carrie A. Hall (2025) [10] provided an extensive overview of body surface area estimation and its clinical applications, underscoring the persistent uncertainty surrounding formula selection. El Edelbi et al. (2021) [11] evaluated multiple regression-based equations for neonates and children, revealing substantial variability among paediatric formulas. Alvear-Vasquez et al. (2023) [12] further demonstrated that BSA correlates with developmental status, emphasizing the physiological importance of accurate estimation. Likewise, Piecuch et al. (2024) [13] compared 25 distinct BSA equations and confirmed that the discrepancies among methods remain significant, even in contemporary practice. Together, these studies underline the ongoing need for rigorous evaluation and potential standardization of BSA assessment methods. Children represent a particularly vulnerable group in this context. Paediatric patients exhibit dynamic changes in body proportions and composition, making adult-derived formulas unreliable when applied to younger populations. The lack of paediatric-specific validation and the diversity of available equations can therefore lead to significant dosing errors, posing potential risks in pharmacotherapy and clinical decision-making.

The aim of the present study is not to introduce a new BSA formula, but to systematically compare existing ones, demonstrate the magnitude of discrepancies among them, and highlight the potential medical consequences of these inconsistencies in paediatric populations. By drawing attention to these issues, we seek to raise awareness of the urgent need for the development of new, validated formulas capable of providing more accurate and consistent BSA estimations in children.

2. Current State of Knowledge

Interest in accurately estimating human Body Surface Area (BSA) dates back to the late 19th century. The earliest attempts, such as Meeh’s 1879 equation [14], relied solely on body weight, reflecting the limited anthropometric data available at the time. A major methodological breakthrough occurred in 1916 when Du Bois and Du Bois introduced their now-classic height–weight model [4]. Using physical measurements of bandaged subjects, they derived empirical coefficients through regression analysis, thereby creating one of the first formulae linking body size to physiological function. Since then, numerous researchers have sought to refine these approaches, leading to a proliferation of BSA equations differing in structure, underlying data, and target populations. Some studies focused on refining constants within the classical equations, while others developed new models incorporating additional anthropometric variables, such as age, body composition, or ethnic background [15,16,17,18]. The Boyd (1935) model [18] and later formulae by Fujimoto et al. (1968) [15] and Haycock et al. (1978) [19] exemplify this evolution toward population-specific adjustments and nonlinear relationships between height and weight. Mathematically, four primary structural types of BSA equations can be distinguished:

$$\mathrm{BSA}={\mathrm{a}}_0·H^{a_1}·W^{a_2}$$

(1)

$$\mathrm{BSA}={\mathrm{a}}_0·W^{{a_1-a}_2·W}$$

(2)

$$\mathrm{BSA}={\displaystyle \frac{a_0·W+a_1}{W+a_2}}$$

(3)

$$\mathrm{BSA}=a_0·W+a_1$$

(4)

where H means height, W is body weight, and a₀, a₁ and a₂ are estimated parameters (generally long series of numbers). Despite their apparent simplicity, these formulas yield substantially different results because their coefficients were derived from small, often non-representative samples measured with varying techniques and instrumentation. The original Du Bois formula, for example, was developed from measurements of only nine subjects—eight adults and one child—limiting its applicability to diverse paediatric populations. Later formulae, though more data-driven, often used region- or age-specific datasets (e.g., Japanese infants, Indian children, or Western adults), which introduces systematic bias when applied outside those populations. This methodological diversity explains the discrepancies noted in clinical BSA estimation. Each formula is grounded in slightly different assumptions about human morphology and scaling laws. Variations in the exponents applied to height and weight reflect differing hypotheses about how surface area scales with mass and linear dimensions. Consequently, when formulas derived from one demographic or body-type distribution are applied to another—particularly to children, whose proportions change rapidly with growth—BSA estimates may diverge considerably. These differences, though numerically modest, can lead to clinically meaningful dosing errors in pharmacotherapy or fluid resuscitation.

A summary of the most widely used paediatric BSA equations is presented in

Table 1. A summary of detailed information on the most common methods of determining the body surface area (BSA) ratio for children. W indicates weight in kilograms, and H indicates height in centimetres.

No.	Authors	Year	Formula	Reference
1	Meeh	1879	0.123 × W2/3	[14]
2	Rubner & Heubner	1898	0.119 × W2/3	[21]
3	Lissauer	1903	neonates: 0.103 × W2/3 toddlers: 0.1 × W2/3	[16]
4	DuBois & DuBois	1916	0.007184 × W0.425 × H0.725	[4]
5	Boyd	1935	0.0004688 × (1000 × W) 0.8168−0.0154×log10(1000×W)	[18]
6	Banerjee & Bhattacharya	1961	0.007 × W0.425 × H0.725	[22]
7	Costeff	1966	${\displaystyle \frac{4\times W+7}{W+90}}$	[17]
8	Fujimoto et al.	1968	infants: 0.009568 × W0.473 × H0.655 1–5 years: 0.0381 × W0.423 × H0.362 over 6 years: 0.008883 × W0.444 × H0.663	[15]
9	Haycock et al.	1978	0.024265 × W0.5378 × H0.3964	[19]
10	Meban	1983	0.00064954 × (1000 × W)0.562 × H0.32	[23]
11	Mosteller	1987	$\sqrt{{\displaystyle \frac{W\times H}{3600}}}$	[24]
12	Vaughan & Litt	1987	0.02 × W +0.4	[25]
13	Current	1998	0.03433 × W + 0.1321	[26]
14	Nwoye & AlShehri	2005	Neonates: 0.03614 × W0.529 × H0.294 Neonates (simplified): 0.042 × W + 0.074	[20]

, which lists representative formulas from Meeh (1879) [14] to Nwoye and Al-Shehri (2005) [20]. All rely on easily measurable anthropometric parameters such as H and W, ensuring practical usability in clinical settings. However, because they simplify complex three-dimensional body geometry into two-variable models, their results inherently depend on the chosen functional form and constants.

Understanding this diversity is essential for interpreting the results presented later in this paper. As highlighted by reviewers, insufficient methodological clarity regarding data sources, formula selection, and computational procedures can obscure the reproducibility of comparative studies. This section therefore establishes the conceptual foundation for our subsequent analysis: it demonstrates why different BSA formulas yield divergent estimates and clarifies the historical and mathematical reasons behind these inconsistencies. By situating our work within this historical and theoretical framework, we emphasize that discrepancies among BSA formulas are not random, but rather systematic consequences of differing empirical assumptions. Recognizing these origins is a necessary step toward the development of standardized, validated approaches for paediatric BSA estimation.

3. Methods and Applied Software

This study was designed as an in silico comparative analysis of the most frequently used formulas for estimating body surface area (BSA) in children. The aim was not to propose a new equation, but to demonstrate the extent of discrepancies between existing formulas and to emphasize their potential clinical implications, particularly in drug dosing.

Height and weight values were obtained from internationally recognized paediatric growth standards, including the World Health Organization (WHO) and Centers for Disease Control and Prevention (CDC) percentile charts, covering children aged 1–18 years [27]. For each age group, data from several percentile levels (3rd, 10th, 25th, 50th, 75th, 90th, and 97th) were used to reflect the variability typical of real paediatric populations. This standardized approach allowed us to compare formulas objectively, without the influence of empirical sampling biases. Twelve classical and contemporary BSA formulas were selected based on their widespread use in clinical practice and representation of different mathematical types, including weight-based power models [14], height–weight power models [4,19], logarithmic models [18], and linear regression approaches [15,24]. These equations were applied to identical anthropometric datasets, and the resulting BSA values were compared to evaluate the extent of formula-dependent variability.

All calculations were executed on a computer equipped with an Intel Core i7-3770 CPU (Santa Clara, CA, USA); the analysis was conducted using OriginPro 2020 software, which was also used to generate the figures. The Children BSA Calculator was designed employing Borland C++ Builder 6 software [6]. In our analysis, we did not apply formal statistical hypothesis testing to compare the results of different BSA formulas. This decision was intentional, as the discrepancies observed are not the result of random variation but stem directly from the inherent mathematical structures and assumptions of each formula. Therefore, hypothesis testing would not provide additional insight into the source of these differences. Instead, we employed descriptive statistical measures to illustrate and quantify the extent of variability across formulas. The standard deviation (σ) was calculated as

$$\sigma =\sqrt{{\displaystyle \frac{1}{n}} \sum _{i=1}^n{\left(x_{i }-\underset{\_}{x}\right)}^2}$$

where x is the BSA value for the i-th formula, and $\underset{\_}{x}$ is the mean BSA value. The relative standard deviation (RSD), expressed as a percentage, was calculated as

$$\mathrm{RSD} (\%)={\displaystyle \frac{\sigma }{\underset{\_}{x}}}\times 100$$

The maximum discrepancy parameter (DSPmax), reflecting the largest observed difference between formulas, was calculated as

$${DSP}_{max}=max\left(x_i\right)-min\left(x_i\right)$$

These parameters allow for a meaningful description of the magnitude of discrepancies and their clinical implications, without relying on formal hypothesis testing. We believe this approach is more appropriate for the aims of this study, which are to emphasize the existence of clinically relevant divergences between formulas and to highlight the necessity of developing new, more reliable BSA equations.

4. Results

Figure 1 shows the result of the BSA calculations for a twelve-year-old child, employing the formulas from Table 1, which takes into account only height (Meeh, Rubner & Heubner, Boyd, Costeff, Vaughan & Litt, Current). These graphs show how the value of the BSA parameter changes depending on the body weight and height of a patient with a height ranging from 140 to 170 cm and a weight ranging from 30 to 70 kg (a typical range of parameters for this age group determined on the basis of centile growth chart). Whether introducing mass as an additional parameter to the BSA formulae improves their precision is shown in Figure 2.

The expanded discussion of our 12-year-old patient serves to show in a concrete, particular example (and from a statistical point of view) the discrepancies caused by a choice of two totally different methods of BSA determination in a certain age group and the practical consequences this has for therapy. It is not possible to generalise this information, because the analysis of the curves in Figure 3 clearly shows that, for example, some methods will be cut off for a 12-year-old with 150 cm height and 30 kg weight (Rubner & Heubner vs. Vaughan & Litt), other ones for a 12-year-old with 150 cm height and 40 kg weight (Current vs. Vaughan & Litt), and yet other ones for the same child, but with 60 kg weight (Current vs. Banerjee & Bhattacharya). A similar situation will occur with a wide spectrum of parameters, including the age, height and body mass of the child.

To facilitate comparison between individual charts, the BSA axis range is presented on the same scale. It is clearly visible that the choice of the equation to estimate this indicator has a significant impact on the obtained results. The greatest differences between formulas independent of the patient’s weight can be observed in the case of Costeff and Current formulas. Ignorance of these differences will inexorably lead to wrong medication dosages, which, as mentioned in the Introduction, may have tragic consequences, especially in the case of children.

In making an analysis, we want to stress that the key to reducing the dire consequences resulting from BSA value discrepancies lies not in the mean arithmetic value, but in the difference in BSA resulting from replacing one method with another. The direction of error (upwards or downwards) is also of importance. For example, let us consider a case of a 12-year-old child that has 150 cm height and 50 kg body weight. In this case, we chose the optimal method to determine BSA—assume it is the Rubner and Heubner method (BSA = 1.615 m²)—which we later swapped for another method, e.g., Fujimoto et al., (BSA = 1.398 m²). The error (change downwards by 0.217 m²) equals 0.217/1.615 ∗ 100%, that is, 13.4%. If we assumed that it is the method of Fujimoto et al. that should be optimal in the case of our patient, the error (change upwards by 0.217 m²) would be 0.217/1.398 ∗ 100%, that is, 15.5%, which compared to the previous scenario results in a more serious estimation error with all its clinical consequences.

Introducing mass as an additional parameter to the BSA formulae improves their precision, as shown in Figure 2. The dispersion of BSA values in the examined range (identical to the data in Figure 1 for formulae based on patients’ height and weight (Du Bois & Du Bois, Rubner & Heubner, Fujimoto et al., Haycock et al., Meban, Mosteller) is clearly smaller than in the case of equations using only their height. It is worth recalling here that the precision of the formulae used is not the same as their accuracy and does not necessarily mean that the results obtained are close to the actual BSA values. Based on the collected data, it can be concluded that the patient’s body weight has a significantly greater impact on BSA patterns than their height. This implies the need to verify the correctness of determining this parameter in the case of children with above-average body weight.

It is noteworthy that, in practice, the disproportion between the actual and the estimated value may be even greater, because the commonly used formulae are based only on simple anthropometric factors and do not take into account the diversity of the body structure of individual people. Currently, the most frequently used BSA equations are based on the results of experiments on a small group of patients with a standard body structure and do not take into account such situations as dwarfism, physical disabilities or non-standard proportions. This problem is particularly visible in the case of formulae developed for the needs of children’s diagnostics—the still-so-popular formula developed by the Du Bois brothers is based on the measurements of eight adults, and only one child (thus its applicability in paediatric population is highly dubious) [4].

The discrepancies between the BSA formulae from Table 1 (except those intended only for newborns) are shown in Figure 3, which shows the values of this parameter as a function of body weight for a 12-year-old female with a height of 150 cm. It can be observed that the current and Rubner’s & Heubner’s formulae yield results that differ significantly from the values calculated by other equations. In the case of the remaining formulae, the convergence of obtained values is greater; yet, as we move away from the average body weight of a 12-year-old child (marked on the chart with a dotted line), the differences between the estimates of the BSA parameter begin to increase. This applies to deviations in the patient’s body weight both upwards and downwards from the typical value; in the first case, a greater spread of BSA values can be observed. This trend is particularly disturbing in the context of the global trend of overweight and obesity among increasingly younger people—the more obese the population of paediatric patients is becoming, the higher the risk of mistaking dosages or miscalculating their body parameters [28,29,30].

The maximum discrepancy parameter (DSP_max) can be defined as the difference in BSA estimates using the two most different formulae in a given age category. The greatest discrepancies in BSA values for 4-week-old infants were observed for the Banerjee & Bhattacharya and Vaughan & Litt formulas (Figure 4A), while for 12-year-olds the largest discrepancies were found for the Banerjee & Bhattacharya and Vaughan & Litt formulae (Figure 4B). For 4-week-old infants, the minimal DSP_max value in the BSA calculations was about 0.22 m², which corresponds to approximately 80% of the average BSA value for females in this age group. The maximum discrepancies in BSA determination for this case reached 0.28 m², which is over 100% of the typical value of this parameter. The observed disproportions increased as the patients’ height decreased and were, to some extent, dependent on their body weight. In contrast, for 12-year-old patients, the minimum of DSP_max recorded was 0.15 m², representing less than 11% of the average BSA index for females in this age group. The maximum DSP_max value for a 12-year-old was 0.75 m², which corresponds to approximately 54% of the average BSA. In this age group, the largest discrepancies were observed in case of overweight children, showing no dependence on the patient’s height. This phenomenon can be explained by the fact that in the case of 12-year-olds, the greatest discrepancies in determining BSA were observed for formulae that used only the patient’s body weight.

Detailed test results for children from the age groups of 4 weeks and 1, 2, 7, 12, and 17 years are presented in

Table A1. A summary of statistical data on the calculation of BSA using various formulae for children of different age groups.

AGE	PERCENTILE	Weight (kg)		Height (cm)		Average BSA (m2)		σ (m2)		DSPMAX (m2)		RSD (%)		Relative DSPMAX (%)
		Girls	Boys	Girls	Boys	Girls	Boys	Girls	Boys	Girls	Boys	Girls	Boys	Girls	Boys
4 WEEKS	3	3.1	3.4	49.7	50.7	0.233	0.243	0.066	0.065	0.396	0.403	28.45	26.67	169.82	166.00
	25	3.7	4.0	52.1	53.1	0.258	0.270	0.064	0.063	0.410	0.417	24.71	23.24	159.15	154.75
	50	4.1	4.4	53.4	54.4	0.274	0.285	0.062	0.061	0.420	0.427	22.80	21.54	153.40	149.58
	75	4.5	4.8	54.7	55.7	0.289	0.300	0.061	0.060	0.429	0.436	21.16	20.08	148.40	145.04
	97	5.3	5.6	57	58	0.319	0.330	0.059	0.059	0.447	0.453	18.62	17.79	140.16	137.48
1 year	3	7.1	7.8	69.2	71.3	0.392	0.416	0.051	0.050	0.491	0.506	13.09	12.06	125.19	121.61
	25	8.2	9.0	72.3	74.1	0.429	0.455	0.050	0.049	0.514	0.531	11.59	10.86	119.80	116.57
	50	8.9	9.6	74	75.7	0.452	0.475	0.049	0.049	0.529	0.543	10.92	10.37	116.94	114.37
	75	9.7	10.4	75.8	77.4	0.478	0.500	0.049	0.050	0.545	0.558	10.34	9.94	114.04	111.77
	97	11.3	11.8	78.9	80.2	0.527	0.542	0.051	0.051	0.575	0.586	9.69	9.49	109.19	108.11
2 years	3	9.3	9.8	79.6	82.1	0.471	0.488	0.046	0.044	0.540	0.552	9.67	9.08	114.79	112.93
	25	10.8	11.3	83.5	85.8	0.519	0.536	0.045	0.044	0.571	0.582	8.67	8.25	110.06	108.57
	50	11.7	12.2	85.7	87.8	0.547	0.563	0.045	0.045	0.589	0.607	8.30	7.98	107.64	107.72
	75	12.7	13.1	87.9	89.9	0.577	0.591	0.047	0.046	0.623	0.637	8.06	7.79	107.95	107.93
	97	14.9	15.1	91.8	93.6	0.641	0.649	0.052	0.051	0.693	0.701	8.10	7.84	108.12	108.01
7 years	3	17.8	18.6	113	115	0.747	0.770	0.047	0.049	0.792	0.814	6.26	6.39	105.91	105.68
	25	21.1	21.9	120	121	0.840	0.862	0.058	0.062	0.902	0.932	6.96	7.14	107.38	108.15
	50	23.5	24.4	123	125	0.903	0.927	0.069	0.073	0.992	1.026	7.65	7.89	109.80	110.60
	75	26.5	27.8	127	128	0.980	1.012	0.084	0.091	1.103	1.150	8.61	9.02	112.57	113.68
	97	34.7	37.6	133	134	1.172	1.236	0.134	0.154	1.397	1.499	11.43	12.44	119.19	121.27
12 years	3	30.0	29.7	141	139	1.085	1.074	0.097	0.096	1.237	1.225	8.93	8.95	113.97	114.05
	25	37.4	37.1	149	148	1.263	1.254	0.139	0.138	1.506	1.495	10.99	10.98	119.27	119.22
	50	42.8	42.7	154	153	1.385	1.381	0.173	0.173	1.699	1.695	12.48	12.51	122.69	122.74
	75	49.6	50.0	158	158	1.531	1.538	0.219	0.222	1.938	1.952	14.30	14.45	126.61	126.90
	97	66.7	69.8	167	167	1.872	1.930	0.348	0.373	2.528	2.632	18.57	19.33	135.00	136.40
17 years	3	43.9	50.8	154	165	1.407	1.569	0.181	0.222	1.737	1.985	12.85	14.17	123.49	126.50
	25	50.8	60.3	161	173	1.560	1.770	0.226	0.289	1.982	2.318	14.49	16.30	127.07	130.96
	50	55.7	66.9	165	178	1.663	1.903	0.260	0.337	2.153	2.546	15.64	17.73	129.46	133.81
	75	61.9	74.8	169	182	1.789	2.056	0.306	0.399	2.368	2.817	17.08	19.39	132.35	136.99
	97	77.5	93.9	176	190	2.088	2.406	0.429	0.557	2.900	3.461	20.53	23.14	138.88	143.82

(see Appendix A). For each age group, the BSA index was calculated using the appropriate formulas from Table 1, taking into account weight and height for individual percentiles from the growth chart. Then, a statistical analysis of the obtained data was performed, including the calculation of the average BSA value, standard deviation (σ), DSP_max, relative standard deviation (RSD) and relative DSP_max. The average value of the BSA index for a typical child in a given age category (50th percentile) ranged from 0.274 m² to 1.663 m² for girls in the age range from 4 weeks to 17 years and from 0.285 to 1.903 for boys in the same age range. Boys showed a higher average BSA value in most age categories than girls, but the opposite was observed in 12-year-olds. This phenomenon can be explained by the earlier onset of puberty in girls, which causes their higher average height and body weight in the mentioned age group. Due to large changes in the BSA parameter depending on the patient’s age, it is more convenient to use RSD and relative DSP_max for comparison purposes. Figure 5 shows the dependence of RSD and relative DSP_max on the patient’s age. In both cases, the obtained curves had a similar shape and did not differ significantly depending on the child’s gender.

It should also be mentioned that there was no correlation between the percentage deviation of each formula and the age of the patients. Equations that, in a given age category, were characterised by a small deviation from the average, could show significant differences in other groups. For this reason, it was not possible to indicate one formula that would be suitable for describing children of different ages. A potential solution to this problem may be the method proposed in the article [15], which involves selecting the appropriate formula depending on the patient’s age. Based on the collected data, it can be assumed that the analysed formulae are most precise for children aged 2 to 7, while the disproportions between the formulae used increase with age. What is particularly disturbing is that the greatest dispersion of estimated results was found in the case of the youngest children (under 1 year of age), which may translate into problems in the treatment of this especially sensitive age group of patients.

Figure 5 is a graphic illustration of the relation between children’ age and the discrepancies of BSA. At the moment of birth, they are the largest comparatively, because the body surface and body mass are very small; thus, even a small absolute change will result in big relative percentage change. After the neonatal period, the RSD% stays the same until adolescence, when due to puberty-related changes the physiognomical characteristics of both sexes begin to differ more from each other. Until puberty, the gender of the child does not seem to affect the BSA. It should also be mentioned that RSD and DSP_max are correlated in the examined age range.

5. Discussion

BSA plays a pivotal role in numerous medical therapeutic processes, such as cancer treatment [1,31,32,33], transplantation [34,35], toxicology [36], dermatology [37,38], the management of nephrotic syndrome [39], and burn treatment [40,41]. Additionally, it is an indicator related to skin water evaporation [42] and is valuable for assessing environmental occupational risks [43]. Several antiviral drugs (e.g., acyclovir), antimicrobials, and antifungals (like caspofungin) necessitate dosing regimens based on the patient’s body surface area. It is believed that medications accumulating in extracellular fluid should be dosed according to BSA, since it reflects the volume of extracellular fluid and of total body water more effectively than body weight. For this purpose, BSA calculators and nomograms are commonly employed [44].

It is obvious that the use of various formulae (in the case of children and adolescents, 14 different formulas are known) when calculating this parameter leads to discrepancies in the results obtained. In the studied cases, the minimum of DSP_max in BSA calculations was approximately 0.22 m², equivalent to about 80% of the average BSA value for females in this age group. The maximum discrepancies in BSA determination in this case reached 0.28 m², exceeding 100% of the typical value for this parameter. The observed disproportions increased as the patient’s height decreased and were, to some extent, dependent on their body weight. In contrast, for 12-year-old patients, the noted minimum of DSP_max was 0.15 m², representing less than 11% of the average BSA index for females in this age group. The maximum DSP_max value for a 12-year-old was 0.75 m², corresponding to approximately 54% of the average BSA. These results are even larger than an extremum noted previously for adults, of about 33% [5].

In order to better illustrate the described problem, a list of doses of drugs used in oncological therapy in children depending on the value of the BSA parameter was prepared (see

Table 2. Discrepancies between dosage of oncological drugs in a 12-year-old girl with a height of 154 cm and a weight of 42.8 kg depending on the value of the BSA parameter.

Disease	Regimen	Reference	Drug	Therapeutic Singular Doses	Dosage			Comments/Assumptions
					MIN	AVG	MAX
Wilms’ tumour [45,46]	EE4A	[45]	vincristine	1.5 mg/m2	1.884 mg	2.078 mg	2.4 mg
			dactinomycin	1.35 mg/m2	1.696 mg	1.87 mg	2.161 mg
	DD4A		vincristine	1.5 mg/m2	1.884 mg	2.078 mg	2.4 mg
			dactinomycin	1.35 mg/m2	1.696 mg	1.87 mg	2.161 mg
			doxorubicin	40 mg/m2	50.24 mg	55.4 mg	64.04 mg
	M	[47]	vincristine	1.5 mg/m2	1.884 mg	2.078 mg	2.4 mg	dosage of dactinomycin given in mg/kg
			dactinomycin	0.045 mg/kg	1.935 mg
			doxorubicin	45 mg/m2	56.52 mg	62.325 mg	72.045 mg
			cyclophosphamide	440 mg/m2	552.64 mg	609.4 mg	704.44 mg
			etoposide	100 mg/m2	125.6 mg	138.5 mg	160.1 mg
Retinoblastoma	VEC	[48,49]	vincristine	1.5 mg/m2	1.884 mg	2.078 mg	2.4 mg	unilateral, intraocular retinoblastoma
			carboplatin	560 mg/m2	703.36 mg	775.6 mg	896.56 mg
			etoposide	100 mg/m2	125.6 mg	138.5 mg	160.1 mg
Ewing sarcoma	VDC-IE	[50]	vincristine	1.4 mg/m2	1.758 mg	1.939 mg	2.241 mg
			doxorubicin	75 mg/m2	94.2 mg	103.875 mg	120.075 mg
			cyclophosphamide	1200 mg/m2	1507.2 mg	1662 mg	1921.2 mg
			ifosfamide	9 mg/m2	11.304 mg	12.465 mg	14.409 mg
			etoposide	500 mg/m2	628 mg	692.5 mg	800.5 mg
Neuroblastoma	IrEC	[51]	irinotecan	100 mg/m2	125.6 mg	138.5 mg	160.1 mg	2nd line, proposed for relapse
			etoposide	100 mg/m2	125.6 mg	138.5 mg	160.1 mg
			carboplatin	80 mg/m2	100.48 mg	110.8 mg	128.08 mg
	CEM	[52]	carboplatin	800 mg/m2	1004.8 mg	1108 mg	1280.8 mg	bone marrow conditioning before HSCT in high-risk neuroblastoma
			etoposide	500 mg/m2	628 mg	692.5 mg	800.5 mg
			melphalan	70 mg/m2	87.92 mg	96.95 mg	112.07 mg
ALL	EsPHALL + Berlin Frankfurt Munster	[53]	dexamethasone	20 mg/m2	25.12 mg	27.7 mg	32.02 mg	not all drugs used simultaneously; not all of drugs used in all cycles
			vincristine	1.5 mg/m2	1.884 mg	2.078 mg	2.4 mg
			methotrexate	5000 mg/m2	6280 mg	6925 mg	8005 mg
			cytarabine	2000 mg/m2	2512 mg	2770 mg	3202 mg
			L-asparaginase	25.000 IU/m2	31400 IU	34625 IU	40025 IU
			cyclophosphamide	200 mg/m2	251.2 mg	277 mg	320.2 mg
			imatinib	300 mg/m2	376.8 mg	415.5 mg	480.3 mg
			daunorubicin	30 mg/m2	37.68 mg	41.55 mg	48.03 mg
			ifosfamide	800 mg/m2	1004.8 mg	1108 mg	1280.8 mg
	MTX/6MP	[54]	methotrexate	20 mg/m2	25.12 mg	27.7 mg	32.02 mg	maintenance regimen; different dosage variants
				40 mg/m2	50.24 mg	55.40 mg	64.04 mg
			6-mercaptopurine	50 mg/m2	62.8 mg	69.25 mg	80.05 mg
				75 mg/m2	94.2 mg	103.875 mg	120.075 mg

Note: The values in the table do not constitute a medical recommendation for the dosage of chemotherapeutics. The sole purpose is to document and visualise the dosage differences resulting from dose determination based on different (but dedicated for this purpose) BSA formulae. It is not recommended to rely on the above values in clinical practice.

). The calculations were made based on the example of a 12-year-old girl with a height of 154 cm and a weight of 42.8 kg, for various BSA values (MIN, AVG and MAX from Table A1). The drugs were grouped based on the disease entity and treatment regimen. The tumours selected are diseases that either occur only in children or are most common in this age group. The final column collects comments relating to additional clinical contexts in which the dosing regimen was applicable. The presented discrepancies between drug doses calculated using different formulae best illustrate the importance of the described issue and its impact on the effectiveness of the treatment process for cancer in children.

Inappropriate dosages of oncological cytostatic drugs can have dangerous consequences. All chemotherapeutic medications are known to have side-effects, some of them life threatening. A study by Miller et al. [55] enumerated the most popular error categories in paediatric chemotherapy administration and found that ¼ of them were related to misdosages. Of the drugs from Table 2, side effects are particularly dangerous in case of methotrexate (a risk of acute, life-threatening gastrointestinal bleeding in case of overdose), carboplatin (acute kidney failure), and cyclophosphamide (haemorrhagic cystitis). These phenomena are especially dangerous for underserved populations. While affluent countries can afford the huge costs of treatment of all side-effects (which can sometimes reach higher sums than treating the tumour itself—let us consider, e.g., a heart transplant because of cardiotoxic effects of doxorubicin, or lung transplants because of damaging action of bleomycin), many populations are excluded from such benefits due to lack of proper hospitals or properly qualified medical staff. Not every country can afford a programme of oncological therapy, and especially in the limited setting, dosages are of crucial importance—if we have sparse resources, care should be taken to distribute them as properly and efficiently as possible, avoiding overdosages (not only are overdosages dangerous, they are also inefficient).

6. Conclusions

The analysis we carried out did not allow for a clear choice of the best formula for determining BSA in children; moreover, it showed that the formulae used overlap only to a limited extent. For this reason, we propose either the improvement of existing BSA formulas or the development of a new one, based on a larger number of easily measurable anthropometric parameters. This is particularly important due to large differences in the body condition of individual people, sometimes additionally aggravated by certain disease lesions. An example of a disease causing a significant deviation in BSA value from the average is obesity [56], which affects an ever-increasing part of the population, including children and adolescents [29]. Improved formulae for calculating BSA for adults were proposed in the article [57]; it is nevertheless necessary to derive distinct equations for children. In order to better illustrate and at the same time attempt to solve the problem described in this article, we have developed a free BSA calculator [Supplementary Material] based on the formulae from Table 1. This software allows us to calculate BSA values using various methods and displays disproportions between them.

A limitation of the present study is that the results are derived from simulated calculations using existing BSA formulas, rather than from empirical measurements of children’s body surface area. While this approach enables a systematic comparison across multiple formulas and a broad range of anthropometric data, it does not directly validate the accuracy of these formulas against real-world measurements. Therefore, the discrepancies we describe reflect theoretical differences between equations rather than absolute deviations from actual BSA values. Nevertheless, this limitation does not undermine the overall objective of the manuscript, which was not to identify the single “best” formula, but rather to emphasise that the coexistence of numerous formulas yields substantially different results. These differences, in turn, may lead to inappropriate medical treatment if the variability is overlooked in clinical practice. Future studies incorporating large-scale empirical datasets and direct measurements will be essential to confirm and refine these findings, and ultimately to support the development of new, more reliable formulas for paediatric BSA assessment.