SciELO - Scientific Electronic Library Online

vol.117 issue5Nutritional evaluation of children with phenylketonuriaThe effect of chronic nitric oxide inhibition on vascular reactivity and blood pressure in pregnant rats author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Sao Paulo Medical Journal

Print version ISSN 1516-3180On-line version ISSN 1806-9460

Sao Paulo Med. J. vol.117 n.5 São Paulo Sept. 1999 

Original Article

Werther Brunow de Carvalho, Paulo Sérgio Lucas da Silva,
Chiu Seing Tsok Paulo, Marcelo Machado Cunio Fonseca,
Luiz Antônio Belli


Comparison between the Comfort and Hartwig sedation scales in pediatric patients undergoing mechanical lung ventilation
Pediatric Intensive Care Unit, Hospital do Servidor Público Municipal de São Paulo,
São Paulo, Brazil




CONTEXT: A high number of hospitalized children do not receive adequate sedation due to inadequate evaluation and use of such agents. With the increase in knowledge of sedation and analgesia in recent years, concern has also risen, such that it is now not acceptable that incorrect evaluations of the state of children's pain and anxiety are made.
OBJECTIVE: A comparison between the Comfort and Hartwig sedation scales in pediatric patients undergoing mechanical lung ventilation.
DESIGN: Prospective cohort study.
SETTING: A pediatric intensive care unit with three beds at an urban teaching hospital.
PATIENTS: Thirty simultaneous and independent observations were conducted by specialists on 18 patients studied.
DIAGNOSTIC TEST: Comfort and Hartwig scales were applied, after 3 minutes of observation.
MAIN MEASUREMENTS: Agreement rate (kappa).
RESULTS: On the Comfort scale, the averages for adequately sedated, insufficiently sedated, and over-sedated were 20.28 (SD 2.78), 27.5 (SD 0.70), and 15.1 (SD 1.10), respectively, whereas on the Hartwig scale, the averages for adequately sedated, insufficiently sedated, and over-sedated were 16.35 (SD 0.77), 20.85 (SD 1.57), and 13.0 (SD 0.89), respectively. The observed agreement rate was 63% (p = 0.006) and the expected agreement rate was 44% with a Kappa coefficient of 0.345238 (z = 2.49).
CONCLUSIONS: In our study there was no statistically significant difference whether the more complex Comfort scale was applied (8 physiological and behavioral parameters) or the less complex Hartwig scale (5 behavioral parameters) was applied to assess the sedation of mechanically ventilated pediatric patients.
Key-words: Sedation scale. Comfort. Hartwig. Mechanical Lung Ventilation. Pediatric Intensive Care.




One of the most important goals in the treatment within pediatric intensive care units (PICUs) is the management of pain and distress in children receiving artificial ventilation. Controlled studies have demonstrated reduced morbidity and mortality rates in patients when adequate analgesia is provided.1, 2 To maintain low levels of stress, pain and fear, these patients require special attention from nursing staff and parents, but additional pharmacological treatment is also necessary. Therefore PICU staff routinely attempt to control distress by administering sedatives and analgesics, and/or managing the social and physical environment. However, to determine the effect of sedation, many sedation-score scales have been developed in order to reduce the subjective impressions, which lead to wide individual variation in evaluation among the clinical staff.

The Comfort scale3 is based around eight behavioral and physiological parameters developed from literature reviews and surveys among experienced PICU nurses. This empirical scale is a reliable and valid method for assessing children's distress and it is non-intrusive, multidimensional, suitable for continuous observation, and it may include variables that remain variable in the face of the continuously changing state of the patient's disease.

The Hartwig scale4 is a less complex sedation score based on five behavioral criteria. This is another empirical scale developed from surveys among experienced PICU nurses. It was devised to quantify the effect of sedation during routine procedures such as tracheal aspiration and its validity and reliability in the clinical assessment of the degree of sedation in patient populations has already been demonstrated.

We performed a prospective trial comparing these two sedation scales in pediatric patients undergoing mechanical ventilation.



Study population. The study was conducted during an 11-month period from March 1995 to January 1996 in the PICU at Hospital do Servidor Público Municipal (HSPM) in São Paulo, Brazil. To evaluate agreement between these two sedation scales we performed simultaneous and independent ratings conducted by specialist pediatric intensive care physicians using the American Comfort scale (Fig 1) and the European Hartwig scale (Fig 2) in pediatric patients undergoing mechanical ventilation.


5a03f1.gif (83517 bytes)

Figure 1 - The COMFORT scale3 - One point was given for the highest and 5 points for the lowest rate of sedation (range from 8 to 40 points). We considered sedation as excessive in the range 8 to 16, adequate (17 to 26) or insufficient (27 to 40).



5a03f2.gif (54142 bytes)

Figure 2. The Hartwig scale4 - One point was given for the highest and 5 points for the lowest rate of sedation (ranging from 8 to 25 points). We considered sedation as excessive in the range 8 to 14, adequate (15 to 18) or insufficient (19 to 25).


Inclusion criteria. a) age < 18 years; b) mechanically ventilated patients receiving intermittent mandatory ventilation or continuous positive airway pressure; c) patients with endotracheal intubation or with a tracheostomy in place. Each patient was sedated by the managing physician using opiates, benzodiazepines, barbiturates, or a combination of these medications. All patients had continuous cardiorespiratory monitoring, and blood pressure monitoring via an inserted arterial catheter.

Exclusion criteria. a) head injury; b) ischemic encephalopathy; c) stroke; d) mental dysfunction; e) multiple trauma within 72 hours of the study; f) abnormalities of muscle function; g) neuromuscular blockage; h) chronic cough. The exclusion criteria were selected to ensure that normal neurologic responses were being assessed, thus avoiding misunderstanding of items assessed by these two sedation scales (such as muscle tone or response to the ventilator), and to reduce the likelihood of distress due to uncontrolled pain.

Data collection and definitions. Each study consisted of a 3-minute period of intensive observation of the patient in his or her pediatric ICU bed. After each observation, evaluations using Comfort scoring (ranging from 8 to 40) and Hartwig scoring (ranging from 8 to 25) were done by the specialist. We graded the sedation given by Comfort scores as follows: adequate (17 to 26 points), excessive (8 to 16 points) and insufficient (27 to 40 points). Using the Hartwig scores, the corresponding sedation grades were: adequate (15 to 18 points), excessive (8 to 14 points) and insufficient (19 to 25 points).

Statistical methods. We used the agreement rate (kappa) with unitized distribution, and p < 0.01 was considered significant.5



This study comprised 30 observations in 18 mechanically ventilated pediatric patients aged 16 days to 5 years (mean: 16.45 months, SD 17.27; see Table 1) and the reason for PICU admission were: cardiac disease - 1 case (5.5%); neurologic disease - 1 case (5.5%); infectious disease - 7 cases (39%); respiratory disease - 9 cases (50%).


5a03t1.gif (2988 bytes)


The analysis of the degree of sedation in our patients obtained by applying the Comfort and Hartwig scales showed almost the same results among those with adequate sedation (Table 2).


5a03t2.gif (6392 bytes)


On the Comfort scale, the mean scores for adequate, insufficient and excessive sedation were: 20.28 (SD 2.78), 27.5 (SD 0.70), and 15.1 (SD 1.10), respectively. On the Hartwig scale, the average scores for adequate, insufficient and excessive sedation were: 16.35 (SD 0.77), 20.85 (SD 1.57), and 13 (SD 0.89), respectively. The analysis of agreement between Comfort and Hartwig scores is presented in the table 2.



In recent years the administration of sedative and analgesic agents has been widely studied and such agents have been applied in the control of stress in critically ill patients, especially children. Sedation and analgesia are known to be powerful instruments providing comfort and reducing complications.6 The utilization of these drugs has been frequently based on subjective personal evaluation without a valid objective method of measuring the distress.

Although a number of reliable and valid methods have been developed to provide observers with a rating of behavioral and physiological indices for children's distress during hospitalization in PICUs, these scales have specific characteristics that may be questioned. Some scales have been developed from the observation of reactions during painful procedures.7 These scales appear inadequate for continuous observation because of the stress factor during the nociceptive stimulation procedure. In addition, other objective scales are inapplicable in pediatric intensive care units because they do not evaluate neonatal or critically ill patients.8

In order to get more experience in the application of objective sedation scales and to try to validate a less complex scale with fewer variables, we made a comparison between two distinct methods.

The Comfort scale has previously been shown to be reliable and has been validated as a descriptor of behavioral and physiological distress with good results. However, its application is not easy because of the great number of variables (eight), thus rendering it not very practical. In addition to this, its applicability is questionable when used in a routine manner.

The Hartwig scale measures only behavioral variables, and therefore has an advantage because of its facility of application. The need for endotracheal aspiration is questionable due to the painfulness of this procedure, thus affecting the goal of our study on sedation. Nevertheless, this fact is minimized as it is a routine procedure done in artificially ventilated patients. Both scales were designed to be age-independent and used at any time to assess the adequacy of sedation at that point in time. For this reason repeated observations of the same patient were not excluded.

We noted that there was a low standard desviation for the means considered for the degrees of sedation on both sedation scales. When the results were analyzed, the agreement rate observed was 63%, and it was statistically significant because there was little difference between the scores in spite of the fact that we had a low n in our sample.



There was no statistical difference when the Comfort and Hartwig scales were applied in mechanically ventilated children, therefore allowing their use in daily clinical practice.



1. Anand KJS, Hansen DD, Hickey PR. Hormonal-metabolic stress responses in neonates undergoing cardiac surgery. Anesthesiology 1990;73:661-70.         [ Links ]

2. Anand KJS, Hickey PR. Halothane-morphine compared with high-dose sulfentanyl for anesthesia and postoperative analgesia in neonatal cardiac surgery. N Engl J Med 1992;326:1-9.         [ Links ]

3. Ambuel B, Hamlett KW, Marx CM, et al. Assessing distress in pediatric intensive care environments: the COMFORT scale. J Pediatric Psycho 1992;17:95-109.         [ Links ]

4. Hartwig S, Roth B, Theison M. Clinical experience with continuous intravenous sedation using midazolam and fentanyl in the pediatric intensive care unit. Eur J Pediatric 1991;150:784-8.         [ Links ]

5. Marx CM, Smith PG, Lowrie LH, et al. Optimal sedation of mechanically ventilated pediatric critical care patients. Crit Care Med 1994;22:163-70.         [ Links ]

6. Fleiss JL. Statistical methods for rates and proportions. 2nd ed. John Wiley & Sons; 1981:212-25.         [ Links ]

7. Jay SM, Ozolins M, Elliot CH. Behavioral management of children's distress during painful procedures. Behav Res Ther; 1985;23(5):513-20.         [ Links ]

8. McGrath PJ, Johnson G, Goodman JT, et al. CHEOPS: a behavioral scale for rating postoperative pain in children. In: HL Fields, et al. editors. Advances in pain research and therapy. Vol. 9, 1985;305-402        [ Links ]


Acknowledgments: Dr. Edison Cunha for the help given in the review of the article.

Werther Brunow de Carvalho - Professor and Director of the Pediatric Intensive Care Unit, Pediatric Department, Universidade Federal de São Paulo/Escola Paulista de Medicina.
Paulo Sérgio Lucas da Silva - Assistant Physicians, Pediatric Intensive Care Unit of Hospital do Servidor Público Municipal de São Paulo.
Chiu Seing Tsok Paulo - Assistant Physicians, Pediatric Intensive Care Unit of Hospital do Servidor Público Municipal de São Paulo.
Marcelo Machado Cunio Fonseca - Assistant Physicians, Pediatric Intensive Care Unit, Pediatric Department, Universidade Federal de São Paulo/Escola Paulista de Medicina.
Luiz Antônio Belli - Assistant Physicians, Pediatric Intensive Care Unit, Pediatric Department, Universidade Federal de São Paulo/Escola Paulista de Medicina.
Sources of funding: Not declared
Conflict of interest: Not declared
Last received: 12 February 1999
Accepted: 30 April 1999

Address for correspondence:
Werther Brunow de Carvalho
R. São Paulo Antigo 145, 10º andar,
São Paulo/SP - Brasil - CEP 05684-010




CONTEXTO: É elevado o número de crianças hospitalizadas que não recebem uma sedação adequada, devido a avaliação e ao uso inadequado desses agentes. Nos últimos anos aumentou a preocupação e os conhecimentos a respeito da sedação e analgesia, de tal modo que é inaceitável hoje em dia termos uma criança sem correta avaliação de seu estado de dor e ansiedade.
OBJETIVO: Comparar o uso de duas escalas de sedação (COMFORT e HARTWIG) em pacientes pediátricos em ventilação pulmonar mecânica.
TIPO DE ESTUDO: Estudo prospectivo.
LOCAL: Unidade de cuidados intensivos pediátricos de hospital terciário.
PARTICIPANTES: Realizadas 30 observações, simultâneas e independentes, em 18 pacientes, por médicos especializados.
TESTE DIAGNÓSTICO: Após observação de três minutos, aplicou-se critérios objetivos (escalas de COMFORT e HARTWIG) para avaliar a sedação.
VARIÁVEL ESTUDADA: Taxa de concordância (Coeficiente Kappa) onde p < 0,01 foi considerado significante.
RESULTADOS: Escala COMFORT (média e desvio padrão) para adequadamente sedado, inadequadamente sedado e muito sedado foi 20,28 ± 2,78, 27,5 ± 0,70 e 15,1 ± 1,10 respectivamente. Para a escala de HARTWIG: 16,35 ± 0,77, 20,85 ± 1,57 e 13,0 ± 0,89. Taxa de concordância observada 63% (p = 0,006), esperada de 44% com o coeficiente kappa = 0,345238 (z = 2,49).
CONCLUSÕES: Em nosso estudo, não houve diferença significativa na aplicabilidade entre as escalas de COMFORT e HARTWIG em assegurar o nível de sedação em crianças submetidas a ventilação pulmonar mecânica.
Palavras-chave: Escala de sedação. Confort. Hartwig. Ventilação pulmonar mecânica. Cuidados intensivos pediátricos.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License