MILANS FORGOTTEN EPIDEMIC OF SUMMER 1629, A FEW MONTHS BEFORE THE LAST GREAT PLAGUE: AN INVESTIGATION INTO THE POSSIBLE CAUSE.

An epidemic not attributable to plague caused thousands of deaths in Milan in the summer of 1629, a time of war and famine that immediately preceded the even more fatal Great Plague of 1630, which is estimated to have killed tens of thousands of people. The 5,993 deaths of 1629 recorded in the Liber Mortuorum of Milan (a city with an estimated population of 130,000 inhabitants at the time) were 45.7% more than the average number recorded between 1601 and 1628. Mortality peaked in July, and 49% of the deaths were attributed to acute fever without a rash or any organ involvement. These deaths involved 1,627 males and 1,334 females, and occurred at a median age 40 years (range 0-95). In this paper, we discuss the possible cause of the epidemic, which may have been an outbreak of typhoid fever.


INTRODUCTION
Various historical sources indicate that thousands of deaths in Milan in 1629 were caused by 22 a 'malignant fever' [1][2][3][4][5][6] that was attributed it to a contagion imported by hungry peasants pouring 23 into the city in a search for food [2]. In the spring of that year, people began to drop dead in the 24 street, "some with roots of herbs, others with bran in their mouths" [2]. From the end of May, all 25 beggars were confined to the Lazaretto of St. Gregory, although this was against the advice of the  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint  Sennert's 1608 treatise [14] was used as a reference to interpret the meaning attributed to the 55 different types of fever associated with the deaths by the drafters of the certificates. Accordingly, 56 deaths due to fever or in which fever was involved were categorised as followed: a) fever without a 57 description of organ involvement; b) fever associated with gastrointestinal symptoms; c) fever 58 associated with respiratory symptoms; d) fever with rash; and e) febre etica, a term describing the 59 chronic/long-lasting fever often associated with phthisis or long-lasting/debilitating diseases [14].

60
The deaths attributed to epilepsia, a term that may have been used to describe infantile 61 febrile convulsions or infantile weakness without any associated fever, epilepsy or epileptic disease 62 in people of any age were not included in the analysis of deaths caused by febrile illness.

63
The weekly number of the deaths due to fever was recorded, as were the personal data of the 64 people whose deaths were attributed to the different types of fever (age, sex, and the number of 65 days between symptom onset and death when available), and the name and professional position 66 (non-graduate surgeon or graduate doctor) of the health officer certifying each death.
. CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Finally, the chi-squared test was used to compare the percentage of deaths due to fever 68 without organ involvement in males and females stratified by age.

71
The 1629 MML lists 5,997 cases of death but, as four proved to be duplicates or erroneous

78
The deceased included 3,072 males and 2,788 females; gender could not be deduced in 133 79 cases (2.2%), mainly involving people found dead in the street whose name was unknown. Their 80 median age at the time of death was 40 years (range 0-107); this information was not available in 81 200 cases (3.3%). As shown in Figure 2, there were no significant differences in the distribution of 82 deaths by gender or age group.

83
A total of 3,363 deaths were attributed to, or associated with fever. Table 1,  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted December 6, 2022. ; https://doi.org/10.1101/2022.12.04.22283078 doi: medRxiv preprint 6 91 five cases, two of which were also associated with hemoptysis (cum sputo sanguinis). Febre etica 92 was also reported in the case of two deaths mainly attributed to phtisi. 93 The following definitions of fever without any specific reference to organ involvement were 94 found in the MML: febre acuta (1,793 cases), febre (320), febre longa (315), febre continua (218),  The distribution of deaths due to different types of fever without any description of organ 114 involvement was substantially comparable, with most of the deaths being recorded during the . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted December 6, 2022. ; https://doi.org/10.1101/2022.12.04.22283078 doi: medRxiv preprint 7 115 summer months (Supp. Fig. 1); however, the distribution of the deaths associated with febre etica 116 was not seasonal.
117 Figure 4 shows the number and percentage of deaths due to fever without any organ 118 involvement in males and females stratified by age. The deaths were relatively more frequent in 119 males (53% vs 47.8%; p<0.001), particularly in the stratum 41-45 years (67.3% vs 54.8%; p=0.01). lazarettos and hospitals would be enough to make up the difference. It is therefore likely that the 128 deaths caused by the morbino were perceived as being more numerous than they actually were.

129
Although the chroniclers emphasised the role of migrants in spreading the epidemic, the 130 increase in the number of incident cases of death due to fever started in week 19, and more than 131 80% of the deaths of beggars found dead in the street were recorded between week 24 and week 28, is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted December 6, 2022. ; https://doi.org/10.1101/2022.12.04.22283078 doi: medRxiv preprint 8 140 pathogenesis of the disease causing the deaths, and so it is not surprising that febre continua was 141 more frequently used by the physici collegiati than the two cerusici. However, neither of these types 142 of fever differ from the other types of acute fever in terms of their distribution between the sexes 143 and seasonality, the median age of the subjects at the time of death, or (when this information was 144 available), the duration of the symptom before death.

145
On the contrary, febre etica did not seem to be seasonal and, in comparison with the other 146 types of fever, appeared at a much younger age and a longer time before death. The cases associated  Febre longa was probably a diagnosis of convenience used (mainly by the two cerusici) to 154 reinforce the exclusion of plague, which was believed to be fatal in a matter of days (Sennert, Book 155 IV, Chap. IV, page. 421). Nevertheless, the deaths due to febre longa occurred at a similar median 156 age and had a seasonal peak that similar to that of the large group of acute fevers without organ 157 involvement that caused most of the deaths in 1629. We have therefore included these cases in our 158 analysis of deaths due to fever without organ involvement as a possible part of the epidemic. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted December 6, 2022. ; https://doi.org/10.1101/2022.12.04.22283078 doi: medRxiv preprint 9 163 that was less severe and less lethal than the plague. The most correct and complete formulation 164 would have been febre acuta maligna, which was used almost exclusively by the physici collegiati. 165 The first reported cases of the plague in Milan appeared in October 1629 [1], and it was unlikely 166 that the presence of a large plague epidemic could be hidden for long, particularly from the were experienced in its empirical diagnosis [17]. Taken together, this seems to be sufficient to 171 affirm that Milan's public health authorities were more than capable of recognising a case of 172 plague, and that the morbino of Senatore Settala was something else.

173
Except for the six cases attributable to smallpox, the MML rarely recorded rash or other 174 cutaneous manifestations, which strongly suggests that the morbino was not the epidemic typhus of is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted December 6, 2022. ; https://doi.org/10.1101/2022.12.04.22283078 doi: medRxiv preprint 10 187 [13] suggests that enteric fevers were endemic, and that more or less extensive outbreaks occurred  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted December 6, 2022.  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The authors received no financial support for the research. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted December 6, 2022. ; https://doi.org/10.1101/2022.12.04.22283078 doi: medRxiv preprint  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted December 6, 2022. ; https://doi.org/10.1101/2022.12.04.22283078 doi: medRxiv preprint