67. Surgery

Nature, Causes, and Burden of Surgical Conditions

Surgery is at the end of the spectrum of the classic curative medical model and, as such, has not been routinely considered as part of the traditional public health model. However, no matter how successful prevention strategies are, surgical conditions will always account for a significant portion of a population's disease burden, particularly in developing countries where conservative treatment is not readily available, where the incidence of trauma and obstetrical complications is high, and where there is a huge backlog of untreated surgical diseases (Murray and Lopez 1996). Some surgical procedures can certainly be perceived as forms of secondary or tertiary prevention. Since the publication of the first edition of this book, which did not have a chapter on surgery, the health care community has recognized that the surgical management of some common conditions can indeed be a cost-effective intervention (Javitt 1993; McCord and Chowdhury 2003). The purpose of this chapter is to explore this hypothesis in more depth.

Methods for Determining Burden of Surgical Disease

We have arbitrarily decided to define a surgical condition as any condition that requires suture, incision, excision, manipulation, or other invasive procedure that usually, but not always, requires local, regional, or general anesthesia. We prefer this definition for two main reasons, to one that would define surgery as procedures performed by trained surgeons. First, surgery does not have to be performed by qualified surgeons. Indeed, in developing countries with few doctors, nondoctors can be trained to perform several types of operations satisfactorily. Second, we believe that the concept of surgery should include minor surgical procedures that nurses or general practitioners could perform along with nonoperative management of surgical diseases (for example, certain types of abdominal, thoracic, or head trauma and burns and infections). Any definition of surgery will have limitations, as has ours, and those limitations must be kept in mind when making interpretations, extrapolations, or estimates. Our broad definition is compatible with the concept of regionalized, coordinated, and interdependent services provided at the community clinic level and at the district and tertiary hospital levels. The most difficult task we then face is trying to determine the burden of surgical conditions as measured in DALYs. To our knowledge, this measurement has never been attempted. What we provide here is a starting point, with the understanding that the calculations will change as data are developed.

Our methodology was based on data from the World Health Report 2002: Reducing Risks, Promoting Healthy Life (WHO 2002) and the global burden of disease study (Murray and Lopez 1996). We began by listing all the conditions for which surgery might be indicated into three groups, with group I being communicable diseases, group II being noncommunicable diseases, and group III being injuries. We then undertook a comprehensive literature review for each condition to determine the proportion of the total burden of disease attributable to it and the proportion of the burden that could be prevented or treated by surgery. Essentially, we found no data of value except maybe for cataracts (group II-F), for which a single intervention (intraocular lens removal with or without implant) is or should ultimately be indicated for nearly 100 percent of patients (Dandona and others 1999; Javitt 1993). The World Health Report 2002 attributes 8,269, of a total 1,467,257,000 DALYs, to cataracts (0.56 percent), and all those DALYs are considered potentially surgical. Maternal conditions (group I-C), perinatal conditions (group I-D), diabetes (group II-C), intentional injuries (group III-B), and unintentional injuries (group III-A), to name a few, are much broader categories of conditions for which the demarcation between the surgical and nonsurgical burden is not as clear as for cataracts.

Faced with a near total lack of pertinent data, we decided that the next best approach was to try to obtain consensus on a "best educated guess" for the surgical burden of each condition. We developed a survey instrument that listed all the possible surgical conditions (all potential surgical DALYs representing the maximum imaginable DALYs that could conceivably be surgical). We sent the questionnaire to 32 surgeons in various parts of the world, asking them what was, in their opinion, the proportion of each condition that would require surgery, which we have referred to as estimated surgical DALYs or the conservative minimum. For each of the 18 completed questionnaires, we discarded the two lowest and two highest values for each condition, leaving a sample of 14 surveys. The lowest value of this sample was consistently chosen so as to err systematically on the conservative side. Note that more than 90 percent of all retained values were within 10 percent of the chosen value. We then applied this value to the DALY numbers provided by the World Health Report 2002 for each category of potentially surgical conditions.

Findings

Table 67.1 presents our estimates of the actual surgical burden for each category of potential surgical conditions for the world as a whole and by region. The table indicates that conditions requiring surgery account for a significant proportion of DALYs. Developing more refined, region-specific information to help policy makers will require more detailed data on the burden of surgical diseases (diseases requiring surgical treatment) and on the cost-effectiveness of surgical therapy. To this end, an extremely helpful step would be for international surgical associations to regularly monitor the disease burden attributable to surgical conditions throughout the world.

A few salient points about the burden of surgical diseases can be made from data provided in table 67.1. We estimate very conservatively that 11 percent of the world's DALYs are from conditions that are very likely to require surgery. Our estimated figures are as high as 15 percent for Europe and as low as 7 percent for Africa. Estimated surgical DALYs for the world are 27 per 1,000 population. The estimated figure is about twice as much for Africa (38 per 1,000) as for the Americas (21 per 1,000).

Table 67.2 summarizes the burden of common surgical conditions based on World Health Report 2002 data. A more detailed look at these data allows us to make the following observations:

• Injuries account for 63 million DALYs, or about 4 percent of all DALYs and 38 percent of the world's estimated surgical DALYs.

• Surgical infections, including infected wounds, superficial and deep abscesses, septic arthritis, and osteomyelitis, undoubtedly account for a significant portion of surgical DALYs, but the available data do not permit quantification.

• Surgical DALYs pertaining to acute abdominal conditions, including appendicitis, intestinal obstruction, gastrointestinal bleeding, hernias, and blunt or penetrating injuries also cannot be calculated because of the lack of data.

• Approximately one-third of maternal conditions, including hemorrhage, obstructed labor, and obstetrical fistulas, are surgical, and these represent 10 million DALYs, or 0.7 percent of all DALYs.

• Congenital anomalies refer to an ill-defined grouping of disparate pathologies that includes congenital malformations such as cleft lip and palate, hernias, anorectal malformations, and clubfoot. We estimate that some 50 percent of congenital anomalies are surgical, representing about 14 million DALYs, or 1 percent of all DALYs.

• Malignancies account for 31 million surgical DALYs, or slightly more than 2 percent of all DALYs.

Table 67.3 breaks down the burden of common surgical conditions by region, also showing rates per 1,000 population. The absolute burden of injuries is highest in Southeast Asia, followed by the Western Pacific and Africa. In terms of population rates, whereas injuries account for 10 DALYs per 1,000 population for the world, the estimated figure is almost twice as much for Africa (15 per 1,000) as for Europe (8 per 1,000). Similarly, rates of obstetrical complications are far higher in Africa than elsewhere, at 6 DALYs per 1,000 population. In contrast, Europe has the highest rate of surgical DALYS related to malignancies—9 per 1,000 population.

All these estimates are debatable. Work is needed to obtain more valid, accurate, and reliable data, but in the meantime, we believe that our results represent a conservative and acceptable baseline estimate of the burden of surgical conditions against which prospectively gathered data for given interventions can be compared in order to assess the extent to which they address the burden. In addition, the burden needs to be monitored over time. Evidence suggests that the burden of intentional and unintentional injuries is rising, particularly in Sub-Saharan Africa and the Middle East. Some of the important contributing risk factors include (a) aging populations; (b) increased access to and use of mechanized vehicles and tools without commensurate improvements in roads, traffic control systems, or capacity for trauma care; and (c) persistent armed conflicts (Kaya and others 1999; Krug, Sharma, and Lozano 2000; Meyer 1998; Mock and others 1995; Mock and others 1999; Nantulya and Reich 2002; Peden and Hyder 2002).