SciELO - Scientific Electronic Library Online

vol.54 número3Analgesia peridural continua: análisis de la eficacia, efectos adversos y factores de riesgo para ocurrencia de complicacionesAnestesia en paciente con síndrome de Gilbert: relato de caso índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados


Revista Brasileira de Anestesiologia

versión impresa ISSN 0034-7094versión On-line ISSN 1806-907X

Rev. Bras. Anestesiol. v.54 n.3 Campinas mayo/jun. 2004 



Postoperative sedation at hospital das clínicas, São Paulo, postoperative unit: a retrospective study*


Sedación pós-operatoria en la unidad de apoyo quirúrgico del hospital de clínicas de São Paulo: estudio retrospectivo



Fábio Ely Martins Benseñor, M.D.I; Domingos Dias Cicarelli, TSA, M.D.II; Joaquim Edson Vieira, TSA, M.D.III

ISupervisor da Unidade de Apoio Cirúrgico (UAC), Divisão de Anestesiologia do HCFMUSP; Doutor em Anestesiologia FMUSP
IIMédico Assistente, Divisão de Anestesiologia do HCFMUSP, USP; Especialista em Medicina Intensiva, Associação de Medicina Intensiva Brasileira (AMIB)
IIIAssistente da Divisão de Anestesiologia do HCFMUSP; Doutor em Medicina, Faculdade de Medicina da USP; Professor Colaborador da Disciplina de Clínica Geral da FMUSP





BACKGROUND AND OBJECTIVES: Despite the established benefits of sedatives in postoperative ICUs, there is no agreement on the optimal sedative regimen or the best way to evaluate sedation depth. This retrospective study aimed at evaluating sedative approaches and their effects on ICU stay in our Hospital.
METHODS: Eighty-three continuously sedated patients were studied according to agent and doses used at the following moments: admission-start sedation (TINI), sedation (TSED), end of sedation-extubation (TEXT) and extubation-discharge (TDIS). In addition, ASA physical status and level of sedation according to Ramsay's score were registered. Data were submitted to ANOVA.
RESULTS: Only patients receiving fentanyl were evaluated (N=80). From these, 34 have received another sedative. TINI was 123.4 ± 369, TSED was 852.5 ± 1242.3, TEXT was 241.1 ± 156.6 and TDIS was 1433 ± 1734.4 minutes. There were no differences on sedation doses versus ASA status (p = 0.11). Nevertheless, TDIS was higher in more critically ill patients (p < 0.001). Diastolic pressure and Ramsay score increased during sedation (p < 0.001 and 0.028, respectively).
CONCLUSIONS: Fentanyl, complemented or not by other agents, was adequate for sedation and hemodynamic stability during postoperative intensive care.

Key Words: ANALGESICS, Opioid; fentanyl; INTENSIVE CARE; SEDATION, Venous


JUSTIFICATIVA Y OBJETIVOS: A despecho de los beneficios del uso de hipnosedativos en Unidades de Terapia Intensiva pós-operatorias, no existe consenso sobre régimen de uso o cuantificación de la sedación. Este estudio evaluó el uso de sedativos y sus efectos sobre el tiempo de permanencia en la unidad pós-operatoria del Hospital de Clínicas de São Paulo.
MÉTODO: Ochenta y tres pacientes que recibieron sedación continua fueron estudiados cuanto a los agentes utilizados y respectivas dosis, bien como los siguientes tiempos: admisión-início de la sedación (TINI), sedación (TSED), término de la sedación-extubación (TEXT) y extubación-alta (TALT). Se evaluaron aún la clasificación de la ASA y el nivel de la sedación por la escala de Ramsay. Los datos fueron sometidos a la ANOVA.
RESULTADOS: Apenas los pacientes que recibieron fentanil fueron evaluados (n = 80). De éstos, 34 recibieron otro sedativo. TINI fue de 123,4 ± 369, TSED de 852,5 ± 1242,3, TEXT de 241,1 ± 156,6 y TALT de 1433 ± 1734,4 minutos. No hubo diferencia cuanto a la dosis de sedativos segundo clasificación de la ASA (p = 0,11). Sin embargo, TALT fue mayor en los pacientes más graves (p < 0,001). Presión diastólica y Ramsay se elevaron durante el transcurrir de la sedación (p < 0,001 y 0,028, respectivamente).
CONCLUSIONES: El fentanil, complementado o no por otros agentes, se mostró adecuado cuanto a la calidad de la sedación y estabilidad hemodinámica en terapia intensiva pós-operatoria.




Sedation is integral part of intensive care units (ICU) routine for decreasing discomfort and anxiety associated to such environment. It attenuates stress responses, promotes disagreeable events amnesia, increases tolerance to ventilatory assistance and helps nursing care 1.

On the other hand, inadequate sedation may increase morbidity of ICU patients. Those inadequately sedated may present with psychomotor agitation predisposing to potentially hazardous events, such as the inadvertent removal of catheters or tracheal tubes, or even adrenergic discharge complications such as myocardial infarction 2. Excessively sedated patients may remain for longer periods under mechanical ventilation, increasing the incidence of ventilator-induced pneumonia and pulmonary injuries 3,4.

Consciousness level of ICU sedated patients may vary from total inability to respond to external stimulations, to awaken, though relaxed and cooperative status, allowing adequate interaction with the unit's team. Sedatives administration is titrated according to desired objectives and sedation depth is routinely evaluated through scores taking into account parameters such as consciousness level, agitation, pain and synchronization with the ventilator, among others 2.

Postoperative Support Unit of HC-FMUSP (PSU) is an ICU with 11 beds where patients receive postoperative intensive care after different surgical procedures. Approaches are open and have no impositions, what assures autonomy for the teams made up exclusively by anesthesiologists.

The purpose of this study was to evaluate different sedation approaches adopted by the PSU teams and to establish possible relationships between these approaches and mechanical ventilation time, PSU stay and hemodynamic changes.



After the Hospital's Ethics Committee approval, patients admitted to the PSU and submitted to any type of continuous sedation while under mechanical ventilation, in the period from February to September 2002 were retrospectively evaluated regardless to gender, age or surgery. Exclusion criteria were patients receiving a single sedative dose and those not maintained under mechanical ventilation.

Data were extracted from nursing control sheets and medical prescriptions and evolutions. So, physical characteristics, physical status (ASA) and type of surgery data were obtained for each patient, as well as the need for complementation with other analgesics or sedatives different from those chosen as the first option for sedation of the specific case. Vital parameters, agent's infusion rate and sedation level according to Ramsay's score were recorded every 12 hours.

The following time intervals were also recorded: time elapsed from patient arrival to sedation administration starting (TINI), time during which patient has remained sedated (TSED), time elapsed between end of sedation and extubation (TEXT) and time elapsed between extubation and PSU discharge (TDIS).

Physical characteristics were compared between genders by Student's t test, while One Way ANOVA was used for remaining comparisons involving other parameters (such as need for additional analgesics or ASA classification). Differences above 5% (p < 0.05) were considered statistically significant.



From the 806 patients analyzed, 83 (10.29%) were included in the study for having received some type of continuous sedation until extubation. Three sedation approaches were identified among these 83 patients: continuous fentanyl and midazolam (1 case), continuous propofol (2 cases) and continuous fentanyl (80 cases). Only the fentanyl group was considered for evaluation since the number of patients sedated with propofol or fentanyl-midazolam was not enough for comparisons. Among patients sedated with fentanyl, there have been no significant differences in gender, age and body mass index. Weight and height, however, have differed (Table I).

Among the 80 patients receiving continuous fentanyl sedation, 5 have died while sedated and 2 were transferred, still intubated, to other intensive care units. Thirty-nine patients were sedated with fentanyl alone, while 34 patients needed other sedative agents in addition to fentanyl infusion. Time patients remained sedated (TSED) varied from 90 to 6494 minutes.

Time elapsed from patients arrival to beginning of sedation (TINI) was 123.4 ± 369 minutes (mean ± standard deviation); time patients remained sedated (TSED) was 852.5 ± 1242.3 minutes; time elapsed between end of sedation and extubation (TEXT) was 241.1 ± 156.6 minutes; and time between extubation and PSU discharge (TDIS) was 1433 ± 1734.4 minutes.

In comparing time patients remained sedated (TSED) and time elapsed between end of sedation and extubation (TEXT) between patients of different ASA physical status, no statistical differences were found, although a trend was observed in patients with poorer physical status who had their extubation delayed as compared to ASA I patients. As to time between extubation and PSU discharge (TDIS), patients with worse physical status remained for a longer time in PSU after extubation (Table II).

To allow a more accurate comparison among sedated patients for different periods, total administered dose was equalized for time and weight of each patient. So, a mean fentanyl value was obtained in nanograms per kilo per minute ( administered to each patient during sedation (Table I). This value was then compared between patients with different ASA physical status.

In spite of the trend for more critical patients to receive lower fentanyl doses (27,9 ± 15,2, 25,1 ± 9,3, 21,4 ± 7,1 and 20,3 ± 5,4 respectively for patients classified as ASA I, ASA II, ASA III and ASA IV), such difference was not significant (p = 0.11).

Patients sedated with fentanyl were divided in 4 groups: patients sedated with fentanyl alone (FE, n = 39), patients needing additional sedatives (FE+AD, n = 34), patients who died while sedated (DE, n = 5) and patients transferred still intubated to other intensive care units (TR, n = 2) for whom it was impossible to consider other interval times except for sedation time (TSED). In comparing these groups in terms of mean fentanyl dose (23,1 ± 8,6 in Group FE, 23,8 ± 9,5 in Group FE+AD, 23,2 ± 6,7 in Group DE and 46,5 ± 42,2 in Group TR) there has only been significant difference in Group TR, which has received higher doses as compared to means of remaining groups (p = 0.02).

Sedation time (TSED), time elapsed from end of sedation to extubation (TEXT), and from extubation to PSU discharge (TDIS) were also compared between patients sedated with fentanyl alone and those who needed additional sedatives (Table III).

Heart rate, mean systolic and diastolic blood pressure were evaluated at 12-hour intervals to check sedative therapy effects on cardiovascular system (Figure 1). Only diastolic blood pressure has showed statistically significant increase along time (p < 0.001). Systolic blood pressure and heart rate have not changed during sedation.

Finally, sedation depth according to Ramsay's score was compared along time through analysis of variance for repeated measures (Figure 2) and showed statistically significant increase along time (p = 0.028).



Our study has shown that sedation of patients under mechanical ventilation in intensive care units may be satisfactorily achieved with fentanyl alone. Despite no statistical difference, the association of other sedatives increased mechanical ventilation duration and the time for extubation, as compared to fentanyl alone.

Opioids, barbiturates, benzodiazepines, propofol, ketamine, inhalational anesthetics and a2-adrenergic agonists are some agents used for sedation of intensive care unit patients 2. The availability of so many agents allows for the use of several sedation approaches. A study in 164 USA ICUs has shown that 18 different agents were used for sedation 5.

A task-force of the Society of Critical Care Medicine has published in 1992 a consensus on Intensive Care sedation recommending midazolam or propofol for short-term sedation (up to 24 hours) and lorazepam for long-term sedation (above 24 hours). Haloperidol was reserved for delirium 6. However, in the clinical practice, it could be observed that such recommendations were not followed by most ICUs. A survey presented in the SCCM meeting of 1998 showed that most Units have not adhered to the above recommendations and maintained their own sedation protocols 7.

In Brazil, a study published by the Brazilian Association of Intensive Medicine has shown that fentanyl is the most popular drug for ICU sedation 8. A recent retrospective study in the Surgical Intensive Care Unit, Escola Paulista de Medicina, has shown that fentanyl as a single drug was used in 58% of sedated patients, while the association of fentanyl and midazolam was used in 21.64% of cases 9. In the PSU, HCFMUSP, this preference was also observed. From 83 studied patients, 80 were initially sedated with fentanyl and, from those, 34 have received complementation with other sedative agent, including midazolam, haloperidol and propofol. One could assume that the need for other agents association could be due to the use of lower fentanyl doses than those really needed. However, no difference has been found between mean fentanyl dose administered to patients of both groups, allowing the conclusion that the association was due to a more intensive response of such patients to surgical aggression. Regarding the sedation time, time between end of sedation and extubation and time between extubation and discharge, there has been a trend toward shorter intervals among patients sedated with fentanyl alone as compared to those receiving additional sedatives, however without statistically significant difference (Table III).

Considering that sedation with fentanyl alone has been effective for almost half the patients under mechanical ventilation and that no difference was found between fentanyl doses administered to those patients and to those requiring additional sedatives, one may propose that sedation with fentanyl as sole agent or associated to other drugs should be a routine for patients under mechanical ventilation. It has to be stressed that this is exactly the approach suggested to anesthesiologists on duty in our Unit.

The maintenance of adequate hemodynamic pattern with the use of fentanyl as sole agent also supports its use as intensive care unit sedative drug. Agents such as propofol, inhalational anesthetics and a2-adrenergic agonists are known for inducing more cardiovascular depression, often bad tolerated by patients 10. In addition, fentanyl's analgesic properties may play important role in its success, especially when pain represents a potential stress factor.

There are as many sedation quality evaluation methods as there are sedation techniques. Since first Ramsay's score description in 1974, several sedation scores have been used in clinical studies 11. Most are based on rating items such as consciousness level, psychomotor agitation, presence of pain or synchronization with the ventilator. Although these classifications allow for efficient data collection and standardization, they might be inadequate when different conditions are evaluated in a same item, such as consciousness level and agitation 12. Ramsay's score has been initially used to describe sedation level with alphaxalone-alphadolone and since then it has been used as sedation evaluation tool for different schemes. A review of 31 randomized controlled studies where different sedatives were compared has shown that Ramsay's score has been used in 20 of them 10. Ramsay's score has a good intra-observers reliability and has been validated with regards to Glasgow's Coma Score modified by Cook and Palma, and to Sedation-Agitation Score (SAS) 11-13. So, Ramsay's score is adequate to titrated sedative agents administration to ICU patients. It is interesting to note that there has been an increase in this score along time in our study, which may suggest the development of tolerance to prolonged and continuous opioid administration 14. Another possibility is that clinical improvement has reflected in increased Ramsay's scores.

Data presented suggest that immediate postoperative sedation with fentanyl is satisfactory regarding sedation, quality, hemodynamic stability and time of mechanical ventilation. The association of other drugs, seems to increase mechanical ventilation time and should be avoided whenever possible. Prospective studies are needed to confirm this hypothesis.



01. Mazzeo AJ - Sedation for the mechanically ventilated patient. Crit Cre Clin, 1995;11:937-955.        [ Links ]

02. Benseñor FEM, Cicarelli DD - Sedação e analgesia em terapia intensiva. Rev Bras Anestesiol, 2003;53:680-693.        [ Links ]

03. Cook DJ, Walter SD, Cook RJ et al - Incidence of and risk factors for ventilator-associated pneumonia in critically ill patients. Ann Intern Med, 1998;129:433-440.        [ Links ]

04. Meade MO, Cook DJ, Kernerman P et al - How to use articles about harm: the relationship between high tidal volumes, ventilating pressures and ventilator-induced lung injury. Crit Care Med, 1997;25:1915-1922.        [ Links ]

05. Hansen-Flaschen JH, Brazinsky S, Basile C et al - Use of sedating drugs and neuromuscular blocking agents in patients requiring mechanical ventilation for respiratory failure. A national survey. JAMA 1991;266: 2870-2875.        [ Links ]

06. Shapiro BA, Warren J, Egol AB et al - Practice parameters for intravenous analgesia and sedation for adult patients in the intensive care unit: an executive summary. Crit Care Med, 1995; 23:1596-1600.        [ Links ]

07. Rhoney DH, Murry KR - A national survey of the use of sedating and neuromuscular blocking agents in the intensive care unit. Crit Care Med, 1998;26:A24.        [ Links ]

08. Moritz R, Goldwasser R - Uso de sedativos em unidades de terapia intensiva brasileiras. Rev Bras Terap Intens, 1999;11: 139-145.        [ Links ]

09. Rodrigues Jr GR, Amaral JLG - Experiência clínica com o uso de sedativos em terapia intensiva. Estudo retrospectivo. Rev Bras Anestesiol 2002;52:747-755.        [ Links ]

10. Ostermann ME, Seiferling RA, Keenan SP et al - A systematic review of the evidence supporting the use of sedative agents in the intensive care units: which agent should be used? JAMA, 2000;11:1451-1459.        [ Links ]

11. Ramsay M, Savege T, Simpson B et al - Controlled sedation with alphaxalone-alphadolone. Br Med J, 1974;22,656-659.        [ Links ]

12. De Jonghe B, Cook D, Appere-De-Vecchi C et al - Using and understanding sedation scoring systems: a systematic review. Intensive Care Med, 2000;26:275-285.        [ Links ]

13. Gimeno G, Alcolea M, Arana M et al - ¿Son todas las escalas de sedacion igual de útiles para la valoración de enfermería? Eferm Intensiva, 1999;10:3-12.        [ Links ]

14. Malan Jr TP - Opioid Pharmacology: New Insights and Clinical Relevance, em: Schwartz AJ, Matjasko MJ, Otto CW. ASA Refresher Courses in Anesthesiology, 2000;28:109-119.        [ Links ]



Correspondence to
Dr. Joaquim Edson Vieira
Av. Dr. Arnaldo 455, Sala 2354
01246-903 São Paulo, SP

Submitted for publication May 26, 2003
Accepted for publication August 14, 2003



* Received from CET/SBA da Faculdade de Medicina da Universidade de São Paulo (FMUSP)

Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons