Oncological surgery is associated with the occurrence of complications, many of them are unexpected. Therefore, the one reason for admitting a patient with cancer to an intensive-care unit (ICU) is postoperative recovery.1 In patients with cancer, the postoperative period continues to be a challenge, even with the advanced technology available in ICUs today.2 The principal characteristic of a patient admitted to an ICU is the potential or actual severity of his or her state of health, which may require specialized medical and nursing care, as well as specific care requirements that, contrary to those provided by other hospital units, are only available in an ICU.
The complexity of care, therapeutic management, and the rates of morbidity and clinical severity can be evaluated by systems developed to objectively quantify these parameters. Using standardized terminology, the severity indexes permit several types of analysis, among them are the stratification of patients according to the severity and prognosis of their illness, the establishment of minimum prerequisites for ICU admission, follow-up of the clinical course and response of the patient to the therapy instituted, and comparison of the clinical course of similar patients submitted to different treatments. In addition, these systems enable an evaluation of a specific ICU, and comparisons can be drawn between different services, comparing, for example, expected and actual mortality rates. Indirectly, the cost/benefit ratio of certain procedures for patients in various stages of the disease can also be evaluated.3
Among the indexes in use today, one of the most widely accepted and used all over the world is the Therapeutic Intervention Scoring System (TISS), an instrument that quantifies therapeutic and diagnostic measures and nursing care in ICUs.4,5 The TISS was originally developed by Cullen et al6 of the Massachusetts General Hospital in Boston in 1974 for the dual purpose of measuring the illness severity of the patients and of calculating the corresponding nursing workload in the ICU. The TISS was initially composed of 57 therapeutic interventions that were awarded scores of 1 to 4 according to the time and effort required to perform the nursing activities, the complexity of the interventions carried out, and the patient's requirement for them.6 Because of the therapeutic innovations developed in the field of intensive care, the 1974 version of the TISS was revised and updated by Keene and Cullen in 1984.7 This revised version contained 76 therapeutic actions and was referred to as the TISS-76. It was, however, widely criticized as being extremely time consuming and underwent several modifications to attempt to make it more functional, leading to the development of the TISS-28,8 which comprises only 28 items. Currently, the principal application of the TISS-28 is to evaluate the nursing workload and human resources management in ICUs.9 A validated translation into Portuguese of this instrument was provided for its application in Brazil.10
The demand for beds in ICUs has been growing, but bed availability has not met this growth demand. This fact has led to ICU nonadmission and to the cancellation of elective surgeries11 or to obliging such operations to be carried out without the backup support of an ICU.
There is a scarcity of literature on the complexity of the postoperative care of women who have undergone gynecologic or breast oncological surgery. It was, therefore, decided to carry out an evaluation of this care, using the TISS-28, in an ICU that specializes in gynecology. The objective of this study was to evaluate the mean TISS-28 score in the first 24 hours according to the characteristics of the women, the disease, the clinical syndromes, and the type of surgery. The categories of the TISS-28 that most influenced the score were also assessed. The clinical progress of these women during the postoperative period and the association between TISS-28 and prognosis were also evaluated.
Subjects and Methods
Type of Study and Selection of Subjects
An observational, longitudinal study was carried out between April 22, 2004, and April 21, 2005, in the adult ICU of the Center for Women's Integrated Healthcare of the Universidade Estadual de Campinas. This ICU has 6 beds for the postoperative care of women who have undergone oncological surgery, who are critically ill, or whose state of health is potentially critical. The Center for Women's Integrated Healthcare is part of a university teaching hospital in the city of Campinas, Sao Paulo, Brazil, dedicated exclusively to the integrated healthcare of women. This study was approved by the internal review board of the Universidade Estadual de Campinas, under protocol no. 591/03, and all women who agreed to participate in the study signed the informed consent form before admission. Sample size was calculated at 80 women based on a pilot study of 20 women, with a type I error of 20% and type II error of 5%.
The principal investigator visited the ICU daily to identify cases. All women with a diagnosis of gynecologic or breast cancer, who had undergone gynecologic or breast oncological surgery, and who had been admitted to the ICU in the immediate postoperative period, were included in the study. Women who, despite having been admitted to the ICU within the period described above, remained in the unit for less that 24 hours were excluded from the study. The characteristics of the women, their disease, comorbidities, and the type of surgery carried out were evaluated, and data include age (years), body mass index, oncological diagnosis, presence or absence of cardiopathy, hypertension, diabetes, and pneumopathy, among others. The surgery to which the women had been submitted was also evaluated and classified as gynecologic or breast. All 48 women with gynecologic cancer underwent only gynecologic surgery, whereas of the 35 women with breast cancer, 5 were also submitted to gynecologic surgery for oophorectomy. Other variables analyzed were the duration of the surgery (in hours), whether a blood transfusion was administered (volume and type of blood derivative given), and the occurrence of any complications during surgery.
Instruments
THERAPEUTIC INTERVENTION SCORING SYSTEM-28
The TISS-28 is made up of 7 major categories composed of the following specific items: basic activities; ventilation; cardiovascular, renal, neurological, and metabolic support; and specific interventions.12 The total TISS-28 score ranged from 0 to a maximum of 88 points, a higher score indicating a greater number of therapeutic interventions used, greater illness severity, and a greater nursing workload.13 The total score is divided into 4 classes (I-IV) according to the total score obtained. Class I refers to patients with a score <=19 points, not requiring intensive care. Class II refers to patients with a score of 20 to 35 points for whom ICU care is recommended. Class III refers to patients scoring 35 to 60 points, whose unstable status does require intensive care, whereas class IV comprises those very critically ill patients, scoring more than 60 points, whose admission to an ICU is, therefore, compulsory.14
For the purposes of this study, the TISS-28 included only those therapeutic interventions carried out during the 24-hour period immediately preceding data collection, in accordance with the recommendations of Miranda et al.8 The TISS-28 and the women's clinical data records were completed from the first day of admission to the adult ICU until the women were discharged from the oncological ICU or until their death. The TISS-28 scores corresponding to each day of hospitalization, the date of discharge from the ICU, or the date of death were then registered.
ACUTE PHYSIOLOGY AND CHRONIC HEALTH EVALUATION II
When the patient was admitted to the ICU, she was evaluated by the unit specialist, who calculated her Acute Physiology and Chronic Health Evaluation II (APACHE II) score.15 Included in APACHE II are physiological variables, body temperature, systemic blood pressure, heart rate, respiratory rate, oxygenation, gasometry, serum sodium, potassium and creatinine measurements, hematocrit, leukocytes, and Glasgow Coma Scale. A score <=10 was considered low risk.
Further data were obtained from the patient's medical records; however, and whenever necessary, information was obtained from the ICU nurse or by direct observation of the patients.
Data Analysis
The data collected were entered into an Excel spreadsheet and analyzed using the SAS statistical software, version 8.2. Mean and SD of the TISS-28 scale, measured on the first day the women were admitted to the ICU, were calculated. For the purposes of analysis, the women were placed in groups according to their clinical characteristics, type and duration of surgery, whether a blood transfusion was required, and whether there were any complications during surgery. For each one of the variables, comparison of the mean TISS-28 scores was carried out using analysis of variation or Wilcoxon test in the case of nonnormally distributed data. These same associations were evaluated with respect to death, using Fisher exact test and calculating the odds ratio. The mean TISS-28 score during the first 5 days of admission to the ICU was compared between those women who survived and those who died. Finally, the mean TISS-28 scores were converted into hours of labor according to the type of surgery (according to Miranda et al,4 each point in the TISS-28 score is equivalent to 7.95 minutes of work). The level of statistical significance was established at 5% for all statistical analyses.
Results
Eighty-three women admitted to the ICU during the immediate postoperative period following gynecologic or breast oncological surgery were included in the study. The mean age of these women was 59.3 years (SD = 16.8 years); their mean APACHE II score was 8 points (SD = 4.6) and postoperative death rate was 5% (4/83). The women remained in the hospital for a mean of 6 days (SD = 4.6; median, 4 days) (data not shown in table). With respect to the oncological diagnosis, 48 women (58%) were diagnosed with gynecologic cancer, 11 of whom had endometrial cancer (13%), 22 had ovarian cancer (27%), 14 had cervical cancer, and 1 had vulvar cancer (1%). Of the women with gynecologic cancer, 27 underwent a total abdominal hysterectomy, either radical or not, with or without bilateral salpingooophorectomy, for endometrial or cervical cancer; 16 had an exploratory laparotomy for ovarian cancer; 3 women had a total simple abdominal hysterectomy for cervical or endometrial cancer; 1 woman had vulvar exeresis due to invasive carcinoma of the vulva; and 1 woman had removal of an intraureteral pigtail stent. Thirty-five women (42%) had breast cancer, of whom 2 had a laparotomy because of adnexal tumors. The surgeries carried out in the case of breast cancer were modified radical mastectomy in 27 cases and quadrantectomy in 4 women, 3 of whom had an oophorectomy during the same surgical procedure. Exeresis of the surgical area was carried out in 2 patients (data not shown in table).
Table 1 shows a mean TISS-28 of 19 (SD = 4.3) for the women as a whole (range, 17.2-22.2, according to the different characteristics of the women). Almost half the women (43%) were older than 60 years and 19% were obese. According to their oncological diagnosis, 58% of women had gynecologic cancer and 42% had breast cancer. The mean TISS-28 score was significantly higher in women with gynecologic cancer (P = .01). No differences were observed in mean TISS-28 scores when these were correlated with the presence or absence of comorbidities, such as cardiopathy, hypertension, diabetes, or pneumopathy. Women who underwent gynecologic surgery had a higher mean TISS-28 score than those who had surgery for breast cancer (P = .03). The duration of surgery was shorter than 3 hours in 60% of the women and was not related to the TISS-28 score. Six women required a blood transfusion during surgery, whereas 11 developed complications during the operation. The most frequent complication was bleeding, which occurred in 5 women; 1 woman had cardiorespiratory arrest and another 5 had various other types of complication (urethral lesion, bronchospasm, bladder lesion, and anaphylactic reaction).
Table 2 shows that almost all the women admitted to the ICU received standard monitoring, multiple intravenous medication, daily wound dressing, management of surgical drains, and monitoring of diuresis. Therapy for abnormal pulmonary function, as well as supplementary ventilation support, was required by 68% of women. Thirty-six percent of the women required central venous access, 19% needed mechanical ventilation, and 18% required care with the upper respiratory airways. Specific interventions, unique to intensive care, according to the TISS-28 categories, were provided for 5% of women, whereas 3% received specific interventions following their discharge from the ICU.
Table 3 shows that the only factors significantly associated with death were age and hypertension. All the women who died during the immediate postoperative period were older than 60 years, and 3 were older than 80 years. The occurrence of death during the postoperative period in women younger than 80 years was 1.38%, whereas for women older than 80 years, the mortality rate was 30% (odds ratio = 30.9; CI 95%, 2.8-338.0). One 85-year-old woman who had a radical mastectomy died on the 22nd day following surgery due to liver failure as a result of metastases from her breast cancer. An 83-year-old woman with recurrent cervical cancer who had an ileostomy for intestinal obstruction died on the fifth day following surgery due to sepsis originating in the abdomen. A third woman, 84 years of age, died on the 14th day following total hysterectomy with bilateral salpingooophorectomy for endometrial carcinoma. She suffered a section of the distal urethra during surgery, which was corrected by reimplantation. This patient's illness progressed to multiple organ failure and had a second laparotomy to treat a bowel obstruction on the 14th day following the first operation. She died immediately following the relaparotomy. The fourth death was a 66-year-old woman with advanced cervical cancer and acute bowel obstruction who had an exploratory laparotomy during which a colostomy was created. She died of sepsis on the 18th day following surgery (data not shown in table).
Table 4 shows the follow-up data of the women and their TISS-28 scores during the first 5 postoperative days. Of the 83 women evaluated, 99% remained in the ICU for at least 2 days, 47% remained for at least 3 days, and only 18% remained for 4 days or more. The mean TISS-28 scores of these women during the period they spent in the ICU were generally between 17 and 20 points regardless of the time since surgery. The mean TISS-28 scores the first 3 days following surgery of the women who died were similar to the scores of the women who survived. In only 1 woman did the TISS-28 score increase to 40 and 37 in the fourth and fifth postoperative days, respectively.
The conversion of the TISS-28 score into hours of work showed that women who underwent gynecologic surgery required a mean of 2.7 hours of work during a 6-hour shift, whereas women who had breast surgery, this time, was reduced to 2.3 hours.
Discussion
The mean TISS-28 score was 19 in the population studied. In other words, women who had been gynecologic or breast oncological surgery had only the minimum required score for admission to the ICU. Consequently, most of the women remained only 2 days in the ICU and were then transferred to another hospital unit. With respect to the postoperative period, most of the care given consisted of basic procedures, such as standard monitoring, changing dressings, collecting samples for tests, and taking care of drains. However, it should be emphasized that many of the women required respiratory physiotherapy, ventilation support, and central venous access. This observation suggests that a physiotherapist should be part of the postsurgical therapeutic team of nurses and physicians. In addition, gynecologic and breast surgeons should have some degree of familiarity with the ventilation support system and be competent at performing central venous access.
The quantification indexes of the nursing workload are important instruments for planning and evaluating hospital units, particularly ICUs. Our study shows that the complexity of care required by women who have gynecologic and breast oncological surgery was intermediate and was greater in the case of women who had gynecologic surgery. With respect to the workload, our results show that one nurse could take full responsibility for the care of 2 women who had undergone gynecologic oncological surgery and would still have approximately 20 minutes for carrying out any extra activities in a 6-hour shift. With respect to the postoperative period of women who had undergone breast oncological surgery, if one nurse takes full responsibility for the care of 2 women who have breast surgery, he or she would still have up to 60 minutes for performing any extra activities during a 6-hour shift.
Other studies have shown a correlation between the severity of the disease, mortality, and TISS-28 scores. In this study, however, no such correlation was observed, and the 4 women who died did so after being discharged from the ICU. However, it is to be noted that the 4 women who died after leaving the ICU were significantly older than the mean age of the population studied (except for 1 women aged 66 years) and had extensive disease, which led to bowel obstruction in 3 cases and complications from liver metastases in 1 women. The APACHE II system, used to evaluate the clinical severity of women admitted to the ICU, also failed to correlate with TISS-28 scores, although a correlation has been reported by other investigators.11 On the other hand, APACHE II is known to be a good predictor of the severity of diseases or acute complications. Because death occurred at a late stage in most of the women, the hypothesis could be raised that the complication that led to their death was not present at the time of their admission to the ICU or that the complication had not resulted in systemic repercussion to the woman at the time of her admission when the APACHE II score was calculated.
As a secondary result of this study, a relationship was observed between hypertension and age as possible risk factors for postoperative mortality. It should be emphasized that of the 4 women who died, all were hypertensive and 3 were older than 80 years. Interestingly, the youngest of the deceased women had a TISS-28 score of 40 on day 4 and 37 on day 5, reflecting an understandable tendency toward providing a more intensive care to women at younger ages, despite the disease extension and small likelihood of cure.
Systemic arterial hypertension is a fairly common chronic disease, generally related to advanced age and possibly associated with cardiovascular disease and renal complications.16 In a recent literature review, elderly patients were shown to have a much greater risk at surgery because of the relationship between age and physiological status.17
In conclusion, this study shows that in the postoperative evaluation of gynecologic and breast oncological surgery, the use of the TISS-28 allows quantification of the ICU workload, although this system has not been shown to be a good instrument for evaluating prognosis. Our findings allow us to conclude that the degree of complexity of care during the postoperative period is intermediate and suggest that a multiprofessional team should be formed to provide adequate care for this category of patients. Although there is a recommendation that this type of postoperative care should be provided in an intensive care setting, we realize that this is not always possible in Brazil. General units are a reasonable option to ICUs in most postoperative gynecologic cases, as may be inferred from the TISS-28 values shown in this study. However, to be capable of dealing with demanding patients, personnel must be trained and a multiprofessional team must be formed. The presence of physiotherapists is certainly mandatory because early implementation of exercises unequivocally improves and shortens the postoperative recovery. Besides training, the required number of nurses to match the needs of oncologic patients is certainly superior to that in an ordinary gynecology infirmary, implying a more elevated financial investment, which is not always possible, especially in low-resource settings. On the other hand, a secondary result of this study is the preliminary characterization of a risk group among women who have gynecologic or breast oncological surgery: all those who died were hypertensive and the majority was older than 80 years. Therefore, specific studies are required to adequately evaluate the appropriate treatment of elderly or hypertensive women with cancer. Also importantly, caregivers should be aware of the most likely complications, like postoperative bowel obstruction, which can occur during the ICU stay and are preventable by reducing or eliminating postoperative pain without excessive sedation, which promotes rapid mobilization and return to self-care. Strategies for pain management that reduce postoperative ileus and other adverse reactions to analgesics can be adopted.
References