The only pitfalls, depending on the size of your environment is:
Syslog IPs will need to be updated on all monitored nodes if you are using this (very helpful for NCM change notifications)
SNMP Traps IPs will need to be updated on all hosts (faster notifications and alerts when a node/interface/service goes down)
SNMP whitelist if used will need to be updated on all nodes
ACLs Firewall Rules will need to be updated, especially if locked down on individual hosts and segments of the network.
Other than that you may need to adjust your dependencies, alerts and Maps, as the perspective of your orion server will change completely.
We are getting ready to do the same. May look into a second polling server, email gateway, and email server at the second datacenter. This way you will know even if the primary site is down.