It is necessary to respond to the author reply by Rietman et al.1 In a personal e-mail communication sent before our first letter to the editor2 was written, Dr. Rietman, in regard to the study3 that was the subject of that original correspondence, stated that his group had not performed lymphoscintigraphy before surgery. Now Dr. Rietman and colleagues state the opposite.1 Perhaps this was merely a simple oversight on Dr. Rietman's part. Nevertheless, questions regarding the type of injection technique used as well as whether triangulated skin marking of the body was actually performed also bring up other issues as discussed below.
In their previously published article, Rietman et al.3 described in detail their very well thought out methodology for measuring morbidity. However, it did not provide a real description of the sentinel lymph node biopsy (SLNB) technique itself, a serious issue given the potential impact that the injection method, injection location, type of radionuclide and/or blue dye used, image acquisition technique, and triangulated skin marking of the axilla, among other variables, could have on false-negative rates and morbidity. Only one reference is cited to support the method that was used to perform SLNB.4 This cited reference is a 1998 review article that described the intratumoral radiotracer injection method as an adjunct to the blue dye sentinel lymph node dissection technique that we described in our original correspondence2 and stresses the optimality of the blue dye method. In their reply, however, Rietman et al.1 report the use of a peritumoral injection technique, in clear contrast to the intratumoral method described in the review4 referenced by the original article.3 Nonetheless, this peritumoral injection method described by the authors in their reply1 is itself known to yield a lower sentinel lymph node detection rate when compared with dermal, areolar, or hybrid combination injection techniques.5–7 Furthermore, even if a sentinel lymph node is visualized, it is generally much fainter than lymph nodes visualized using these other injection techniques. Sentinel lymph nodes potentially would be easier to find using these other injection methods, a feature that probably would lead to lower levels of morbidity.
Rietman and colleagues' response concerning lymphoscintigraphy is essentially an indirect quote from the methodology reference,4 which states that lymphoscintigrams are reviewed with the operating surgeon and that the procedure takes place within 24 hours of lymphoscintigraphy.4 This may suggest that the surgeons closely followed the methods described in the methodology reference except for the use of a peritumoral injection technique, as has been noted. If so, this raises other serious issues. The referenced methodology4 uses extensive blue dye dissections as the primary method for finding and harvesting sentinel lymph nodes, with radiotracer use included almost as an afterthought. The investigators describing this methodology state that it involves careful dissection of the lymphatic vessels, after which these vessels are followed to the sentinel lymph node, which can be identified by its blue coloration; subsequently, the radioactivity present in the sentinel lymph node is measured with a gamma detection probe, and if a blue lymph node or blue lymphatics cannot be identified, then the probe is used for tracing. The use of blue dye as a primary method for identifying the sentinel lymph node is a dissection-intensive process, as the blue lymphatic tracts often have to be dissected from a point near the injection site to the axilla before the blue sentinel lymph node is found. The exact way in which blue dye was used by Rietman and colleagues is unclear. However, if the authors followed the referenced methodology as closely as suggested by the description of the lymphoscintigraphy in the reply, then excessively blue dye–driven dissection probably was performed in the SLNB group. When serving as the primary method for sentinel lymph node identification, the use of blue dye works against the goal of morbidity reduction; in contrast, radionuclide-based methods require only a simple, very small incision, just above the sentinel lymph node or at a cosmetically ideal location, in many cases.
Another issue that must be considered is triangulated skin marking of the axilla to inform the surgeon of where the incision should be made. This is a very important consideration, as the sentinel lymph nodes are not always ‘hot’ enough to easily detect with the probe at the skin surface before performing the first incision, particularly when the peritumoral injection method is used. Triangulated body marking from two angles is considered by some surgeons to be as important as, if not more important than, the images themselves. Body marking influences the surgeon as to where to place the initial incision so that he or she can identify the sentinel lymph node with the least amount of dissection, rather than selecting an arbitrary or predetermined incision point and dissecting until a sentinel lymph node is found, thereby increasing the likelihood of morbidity. These markings are especially important when surgery with the probe is performed the day after radiotracer injection, as radioactive decay often makes the sentinel lymph nodes almost 10 times fainter than they are immediately after injection, given the 6-hour half-life of 99mTc. In addition, obese patients often pose a challenge, as the attenuation of radioactivity by overlying soft tissue also diminishes the perceived brightness of the sentinel lymph node at the skin surface.
Did the authors truly intend to cite the report by Rutgers et al.4 as the only description of their technique? This question raises issues that are crucial with regard to the reduction of morbidity, which is, after all, virtually the only reason for performing SLNB.
Dr. Rietman and colleagues note that two of the three surgeons involved in their study had only recently become qualified to perform SLNB. Thirty or more training cases are considered by some to be the minimum requirement before a surgeon can be considered qualified to perform SLNB.8, 9 Rietman and colleagues' own cited methodology reference reported a requirement of 30–50 training cases.4 Although the literature suggests a plateauing in the false-negative SLNB rate after ≥ 30 cases, the question of whether dissection, and therefore morbidity, becomes minimized as surgeons become highly experienced practitioners of this technique has not been addressed. Did all surgeons have equal knowledge regarding optimized lymphoscintigraphy and triangulated skin marking of the axilla, as well as the need to minimize their reliance on blue dye–driven dissections?
At the conclusion of our original correspondence,2 we cited several newly released articles, all employing lymphoscintigraphy, that reported statistically significant correlations between morbidity and SLNB and axillary lymph node dissection. All studies on this subject differ with respect to a number of variables, and consequently, it is very difficult to make comparisons among these studies. Thus, we agree with Dr. Rietman and colleagues that a direct comparison involving their article, or any other article published on the subject to date, would be difficult. Yet, if some common ground does exist, it might be worth noting, as we did in our original correspondence.
We were puzzled by Rietman and colleagues' response1 to our original letter,2 given the abovementioned e-mail communication sent by the principal author. This communication was in itself the reason for our correspondence.2 Hopefully this was only an oversight by the authors.