There are only a handful of antibiotics approved for treatment of mastitis in the United States. Most of them have been available for many years, and it has been many years since any new treatments have been approved. Considering these facts one would think choosing an antibiotic to treat mastitis for any clinical case would be simple, and that the optimal choice probably has not changed much over time. It turns out that finding the optimal choice is not simple, and that the optimal choice has indeed changed over time. There is an excellent review of this topic by Ruegg in the November 2018 issue of Veterinary Clinics, Food Animal Practice. Much of the information in this column comes from that article. Mastitis is defined as inflammation of the mammary gland. Most of the time that inflammation is caused by a bacterial infection. Cows get mild (grade one) or moderate (grade two) clinical mastitis about 85 percent of the time, and severe clinical mastitis only about 15 percent of the time. Severe cases may represent a real emergency and should be treated as such. However, antibiotic treatment choices of the other 85 percent require more scrutiny. Around the time of World War II, almost all cases of grade one and two mastitis were caused by Streptococcus agalactia and Staphylococcus aureus. As a result, antibiotic treatments and treatment protocols were developed to treat these organisms. Indeed, to this day, some antibiotic tubes only list causation by one of these two organisms as an indicated use on the label. As farmers got better at controlling these two pathogens in the 1980s, coliform bacteria became more common isolates from clinical mastitis. More recently it has been shown that the most common isolates from clinical mastitis are organisms that normally exist in the cows’ environment. However, if we count no growth as a result, it turns out the most common cause of clinical mastitis in most countries today is no growth. The next most common organisms are coliforms (mostly E. coli and Klebsiella) or non-ag Streptococci. We should care about this because treatment of clinical mastitis should be directed at cows that are likely to benefit and not to those that are not. This makes economic sense on the farm and represents better antibiotic stewardship. It is also true that the optimal antibiotic choice will vary depending on what is isolated, and that many isolates will not respond to antibiotic therapy no matter what is chosen for treatment. For example, we know that bacteriologic cure rates for Staph aureus are typically 25 percent or less, while treatment for environmental Streps and non-aureus Staphs would be expected to cure 65 to 75 percent of the time. We would also expect 75 percent or more cases caused by E. coli to cure. However, the rate of cure for non-treated E. coli is about the same as for treated. While environmental Streps often respond to treatment, we also know that typically they have a poor spontaneous cure rate and high rate of recurrence if not treated. We also know that some environmental Streps respond better to longer antibiotic treatments, as does Staph aureus. Then there are a whole bunch of organisms that show little, if any response to antibiotic treatment such as yeasts, Prototheca, Mycoplasma, Pasteurella, Trueperella, Pseudomonas, Serratia and others. Part of the reason all of this is so hard is that the appearance of the milk has little to do with the presence or absence of an organism. So, simply because the milk looks bad, it does not mean treatment is likely to work, and if the milk still looks bad after treatment it does not mean the treatment did not work. Even when the milk looks normal after treatment, it is not necessarily the case that the treatment worked because most of the time milk will return to normal, usually in 4-6 days, whether there is an organism present or not. In the past, it was also thought that performing sensitivity analysis on milk cultures could help guide treatment decisions. Today, most experts agree there is little value in sensitivity testing for routine mastitis. It is also true that some cows will respond more poorly even when infected by an organism that should be responsive to antibiotic. According to Ruegg, only 20 to 33 percent of cases are likely to benefit from non-specific antibiotic treatment. Thus, what really matters is the organism, or lack of organism, and the cow. Treatment decisions need to be based on the organism, and protocols should be developed with the input of or by the herd veterinarian. This is why on-farm culture has become so much more popular in recent years. Thus, in today’s world, most cases of mild and moderate cases should not be treated on most farms. The ones that should be treated might benefit from different protocols based on the organism isolated. The choice to treat or not, and what to use to treat is no longer as simple. Jim Bennett is one of four dairy veterinarians at Northern Valley Dairy Production Medicine Center in Plainview, Minn. He also consults on dairy farms in other states. He and his wife, Pam, have four children. Bennett can be reached at bennettnvac@gmail.com with comments or questions.